Intersection passing intelligent decision-making method and device considering surrounding real-time traffic flow

Abstract

The invention discloses an intersection passing intelligent decision-making method and device considering surrounding real-time traffic flow. The method comprises the steps that an automatic driving vehicle is provided with a sensor to obtain state information; placing a vehicle in a structured road in combination with high-precision map information, and constructing a lane-level relative relationship; acquiring the distance to the front intersection according to the vehicle state and the high-precision map, judging whether the distance is greater than a set threshold value, if so, driving according to a normal open road, and otherwise, estimating the surrounding real-time traffic flow; determining required entrances and exits and corresponding lanes according to the passing direction of the front intersection; distinguishing a front vehicle and left and right vehicle groups according to the information; obtaining vehicle group information according to the positions and historical track points of the front vehicle and the left and right vehicle groups; summarizing the current and historical coordinates of the grouped vehicle groups to obtain overall traffic flow information; performing curve fitting according to the track point set to obtain a driving trend curve; the decision-making module selects a passing mode in the intersection according to the surrounding vehicle trend and the exit lane information.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to an intelligent decision-making method and device for intersection traffic taking into account surrounding real-time traffic flow. Background Art

[0002] As a typical scenario for the implementation of artificial intelligence technology, vehicle autonomous driving technology has developed rapidly in recent years. The ability to reasonably deal with highly dynamic and complex traffic scenarios is a key prerequisite for the widespread promotion of autonomous driving technology. Intersection traffic scenes have many roads intersecting, various combinations of roads and signs, and changing traffic signals, which lead to large changes in the traffic patterns of vehicles and pedestrians. This is one of the dynamic and complex scenarios that autonomous vehicles need to overcome.

[0003] In the prior art, when an autonomous vehicle is passing through an intersection, it will select an exit lane in advance based on the lane it enters the intersection, and use this to determine the path it will take, and then mainly consider the interaction with the environment from the perspective of speed. In particular, when the number of entrance lanes and exit lanes are different in the direction of the intersection that the autonomous vehicle is passing through, selecting the exit lane in advance will lead to increased lane change conflicts and frequent merging conflicts, affecting traffic safety and efficiency.

[0004] CN108459588B discloses an automatic driving method, device and vehicle, which determines the path of the vehicle to the target location of the intersection by considering the traffic signal in the direction of the automatic driving vehicle, and applies different acceleration curves to obtain the driving trajectory including speed and time, and then obtains the optimal trajectory decision according to the driving model, and finally arrives at the intersection smoothly. This patent focuses on the speed control of the automatic driving vehicle before entering the intersection, so that the vehicle can arrive at the intersection in a more stable and safe way. It takes less consideration of the complex interactions that may occur in the intersection, such as the intersection of traffic, which restricts the traffic capacity of complex scenes at the intersection.

[0005] CN110471415B discloses a vehicle with an automatic driving mode and its control method and system, which mainly deals with the traffic strategy of the automatic driving vehicle when the traffic light at the intersection changes, especially according to the time when the traffic light changes from yellow to red, combined with the current position information of the vehicle, to determine whether to continue to pass through the intersection. This patent also focuses on the decision of whether to pass through the intersection itself, but gives less consideration to the traffic mode of the automatic driving vehicle in the intersection, and it is difficult to handle the complex and changeable actual intersection traffic scenes.

[0006] CN109582022B discloses an autonomous driving strategy decision system and method, and proposes an autonomous driving strategy decision method for passing through intersections. It mainly considers whether the driving intentions corresponding to the predictions of multiple third-party vehicles around the autonomous driving vehicle overlap with the driving of the vehicle itself, and then determines the driving strategy of the vehicle itself, such as deceleration. This patent analyzes and considers third-party vehicles separately, and relies on the accuracy of the prediction of third-party vehicles. There may be an overly conservative passage process, and the vehicle itself tends to choose unnecessary deceleration or waiting, reducing driving efficiency.

[0007] US20210278231A1 discloses a system and method for following an autonomous driving vehicle using surrounding traffic, which selects multiple nearby vehicles and integrates the driving patterns of these vehicles into a driving traffic trend, and controls the vehicle to follow the trend, thereby achieving collision-free and smooth driving of the autonomous driving vehicle. After acquiring the traffic flow, this patent only coordinates the autonomous driving vehicle with the driving of surrounding vehicles by following to achieve safe and smooth driving. It does not consider whether the driving goal of the vehicle is consistent with that of other vehicles, especially when there are large differences in the driving of different vehicles at the intersection, which may cause the autonomous driving vehicle to be unable to plan a reasonable path according to its own driving goal, reducing traffic efficiency. Summary of the invention

[0008] In response to the above technical problems, the present invention proposes an intelligent decision-making method and device for intersection traffic that takes into account the surrounding real-time traffic flow, which can solve the safety and efficiency problems of traffic in intersections, especially intersections with different numbers of entrance lanes and exit lanes, and reduce lane change conflicts and merging conflicts.

[0009] In order to achieve the above-mentioned purpose, the technical solution of the present invention provides an intelligent decision-making method for intersection traffic that takes into account the surrounding real-time traffic flow, which includes the following steps: S1: obtaining the status information of the self-driving vehicle and the surrounding vehicles through the sensors equipped by the autonomous driving vehicle; S2: placing the self-driving vehicle and the surrounding vehicles in a structured road in combination with the high-precision map information, obtaining the lane information of the self-driving vehicle and other vehicles, and constructing a lane-level relative relationship; S3: obtaining the distance from the self-driving vehicle to the intersection ahead according to the self-driving vehicle status and the high-precision map information, and judging whether the distance is greater than a set threshold, if so, driving on a normal public road, otherwise proceeding to step S4 to estimate the surrounding real-time traffic flow; S4: determining the required entrance and exit as well as the same-direction traffic lane and exit lane information according to the traffic direction of the intersection ahead; S5: screening based on real-time perception information and in conjunction with lane information Select vehicles traveling in the same direction as the vehicle, and then distinguish the front vehicle, the left vehicle and the right vehicle according to the entrance and exit determined in step S4 and the same-direction lane and exit lane information, and in combination with the lane information of the vehicle obtained in step S2; S6: Based on the current positions and historical trajectory points of the distinguished front vehicle, left vehicle and right vehicle, obtain the information of the front vehicle group, the left vehicle group and the right vehicle group relative to the vehicle; S7: By summarizing the current and historical coordinates of the corresponding grouped vehicle groups, obtain the overall driving traffic information of the corresponding grouped vehicle groups; S8: By curve fitting the trajectory point set of each group of vehicle groups, obtain the corresponding driving trend curve, thereby obtaining the driving trend curve of the surrounding traffic of the vehicle; S9: Select the passage mode in the intersection according to the driving trend curve of the surrounding traffic and the exit lane information.

[0010] Furthermore, in step S7, when the historical coordinates are summarized, only the coordinate information of the corresponding vehicle that enters the set threshold range mentioned in step S3 is included.

[0011] Furthermore, in step S8, when performing curve fitting on the trajectory point set, firstly, a local coordinate system is constructed with the vehicle as the center, and the trajectory point sets of different groups of vehicles are converted into the local coordinate system. Then, a curve fitting problem is constructed in this local coordinate system, and the corresponding fitting curve is obtained by solving it.

[0012] Further, in step S8, assuming that the trajectory of the vehicle follows a 4th-order polynomial curve, a point in the trajectory is represented by the following expression: Among them, x i ,y i is the coordinate value of the i-th trajectory point in the vehicle coordinate system, β j (j=0...4) is the fitting parameter; first, all the trajectory points of the same group of vehicles are written in vector form Among them, X is the polynomial matrix of different orders composed of the x coordinates of all trajectory points, For the parameters to be fitted, y is a column vector consisting of the y coordinates of all trajectory points, then the following objective function is established Find the coefficients of a polynomial Then, the starting and ending values ​​of x are set according to the range of the intersection, and the corresponding step size is determined. The corresponding y value is then calculated through a polynomial expression to obtain the corresponding fitting traffic flow trend curve.

[0013] Furthermore, in step S1, the sensors equipped by the autonomous driving vehicle include: lidar, high-definition camera, IMU / GNSS equipment; the status information of the vehicle includes: the position, velocity vector and acceleration vector of the vehicle; the status information of surrounding vehicles includes: the type, position, speed and relative relationship between the other vehicles and the vehicle.

[0014] Furthermore, in step S2, the relative relationship between other vehicles and the own vehicle includes: the front vehicle in the same lane; the rear vehicle in the same lane; the front vehicle in the right lane, the parallel vehicle in the right lane, and the rear vehicle in the right lane; the front vehicle in the left lane, the parallel vehicle in the left lane, and the rear vehicle in the left lane; and when the road in the same direction has more lanes and there are other vehicles on the corresponding lanes, the relative relationship between other vehicles and the own vehicle is further expanded to cover all lanes in the same direction.

[0015] Furthermore, in step S2, if there is overlap between the roads in different directions in the intersection, so that there are multiple roads in the intersection corresponding to the vehicle's position, resulting in unclear correspondence between other vehicles and the vehicle itself, it is necessary to trace back the historical lane information of other vehicles and mark them according to the relative relationship between the last non-intersection lane where the vehicle was before entering the intersection and the lane where the vehicle is located.

[0016] Furthermore, in step S9, the travel mode includes: travel distance, feasible trajectory and selection of exit lane.

[0017] The technical solution of the present invention also provides an intelligent decision-making device for intersection traffic that takes into account the surrounding real-time traffic flow, which includes the following modules: a state information acquisition module: acquiring the state information of the vehicle and surrounding vehicles through sensors equipped by the autonomous driving vehicle; a lane-level relative relationship construction module: placing the vehicle and surrounding vehicles in a structured road in combination with high-precision map information, acquiring lane information of the vehicle and other vehicles, and constructing a lane-level relative relationship; a threshold judgment module: acquiring the distance from the vehicle to the intersection ahead based on the vehicle state and high-precision map information, and determining whether the distance is greater than a set threshold. If so, driving on a normal public road, otherwise it is necessary to estimate the surrounding real-time traffic flow; an entrance and exit and lane determination module: when the threshold judgment module determines that it is necessary to estimate the surrounding real-time traffic flow, the required entrance and exit and the same-direction traffic lane and exit lane information are determined according to the traffic direction of the intersection ahead; a vehicle differentiation module: based on real-time perception information and in conjunction with lane information Filter out vehicles traveling in the same direction as the vehicle, and then distinguish the front vehicle, the left vehicle and the right vehicle according to the entrance and exit determined by the entrance and exit and lane determination module, as well as the same-direction lane and exit lane information, and the lane information of the vehicle obtained in the lane-level relative relationship construction module; Vehicle group grouping module: According to the current position and historical trajectory points of the distinguished front vehicle, left vehicle and right vehicle, obtain the information of the front vehicle group, left vehicle group and right vehicle group relative to the vehicle; Driving traffic information acquisition module: By summarizing the current and historical coordinates of the corresponding grouped vehicle group, obtain the overall driving traffic information of the corresponding grouped vehicle group; Driving trend curve fitting module: By performing curve fitting on the trajectory point set of each group of vehicle groups, obtain the corresponding driving trend curve, thereby obtaining the driving trend curve of the surrounding traffic of the vehicle; Traffic mode selection module: Select the traffic mode in the intersection according to the surrounding traffic trend curve and the exit lane information.

[0018] The technical solution of the present invention also provides a computer-readable storage medium containing a computer program, characterized in that when the computer program is executed by one or more processors, the intelligent decision-making method for intersection traffic that takes into account the surrounding real-time traffic flow as described above is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 is a schematic diagram of an entrance and an exit and corresponding lanes of an embodiment of the present invention;

[0021] Figure 2A and Figure 2B Schematic diagrams showing different driving modes of other vehicles traveling in the same direction as the vehicle;

[0022] Figure 3A and Figure 3B They are shown respectively with Figure 2A and Figure 2B Example diagram of current and historical trajectory points of vehicle groups corresponding to the two modes of travel;

[0023] Figure 4 A schematic diagram of the overall driving trend obtained according to the trajectory points of different vehicle groups of the present invention is shown;

[0024] Figure 5 It is a flow chart of the intelligent decision-making method for intersection traffic taking into account the surrounding real-time traffic flow of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] The first aspect of the present application proposes an intelligent decision-making method for intersection traffic taking into account surrounding real-time traffic flow, comprising:

[0027] 1. Obtain the current status of the autonomous vehicle and the status of surrounding vehicles through the sensors equipped by the autonomous vehicle.

[0028] The sensors used include lidar, high-definition cameras, I MU / GNSS equipment and other types.

[0029] For the autonomous driving vehicle itself, its current state contains multiple key information, such as the vehicle's position, velocity vector, and acceleration vector.

[0030] The status information of surrounding vehicles includes the type, position, speed, and relative relationship between the vehicle and the self-vehicle.

[0031] 2. Combine high-precision map information to place the vehicle and surrounding vehicles on a structured road, obtain lane information for the vehicle and other vehicles, and build lane-level relative relationships based on the lanes where the vehicles are located to confirm the relative relationships between other vehicles and the vehicle. Possible relative relationships include:

[0032] The front vehicle in the same lane, and there may be multiple vehicles, which will have a relative front-to-back position relationship;

[0033] There may also be multiple vehicles behind in the same lane, and there will also be corresponding relative front and rear relationships;

[0034] The vehicle ahead in the right lane, the parallel vehicle in the right lane, and the vehicle behind in the right lane;

[0035] The vehicle ahead in the left lane, the parallel vehicle in the left lane, and the vehicle behind in the left lane;

[0036] If the road in the same direction has more lanes and there are other vehicles in the corresponding lanes, it can be further expanded to the right-right lane, the left-left lane, and so on, until all lanes in the same direction are covered.

[0037] Due to the overlap of roads in different directions at the intersection, there are multiple roads at the intersection corresponding to the vehicle's position. As a result, when the correspondence between other vehicles and the vehicle itself is unclear, it is necessary to trace back the historical lane information of other vehicles. Generally, they can be marked based on the relative relationship between the last non-intersection lane before entering the intersection and the lane where the vehicle is located, so as to ensure a clear judgment of the relative relationship between the vehicles.

[0038] 3. Based on the state of the autonomous vehicle itself and combined with the high-precision map, the distance from the autonomous vehicle to the intersection ahead can be obtained, denoted as d distance_to_crossroad , if d distance_to_crossroad Greater than the set threshold d threshould , then follow the normal public road; if d distance_to_crossroad Less than the set threshold d threshould , then enter the surrounding real-time traffic estimation step.

[0039] 4. The above-mentioned real-time traffic flow estimation process covers several important steps. First, the entrance and exit of the intersection and its corresponding lane information must be determined. Then, vehicles traveling in the same direction as the vehicle are selected based on the real-time perception information and the lane information, and then the real-time traffic flow is constructed. Finally, the decision system intelligently determines the traffic mode of the intersection based on the constructed real-time traffic flow. Each of these steps will be elaborated in detail below.

[0040] 5. Determine the entrance and exit required for the vehicle based on the direction of travel required by the intersection ahead of the autonomous vehicle. Figure 1 As shown, taking the case of an autonomous driving vehicle going straight at an intersection as an example, the lane where the vehicle is located determines the entrance for this intersection. Then, based on the vehicle's need to go straight and the direction of the entrance, the opposite lane can be determined as the exit.

[0041] 6. With the help of high-precision map information, the same-direction traffic lanes can be obtained by considering the entrance and the direction of travel. Figure 1As shown in the figure, at the entrance, there are three straight lanes, from left to right, namely straight lane 1, straight lane 2 and right turn straight lane. It should be noted that since the traffic direction of the left turn lane is inconsistent with the traffic direction of the autonomous driving vehicle, it is excluded from the calculation although it is at the corresponding entrance.

[0042] 7. Using high-precision map information and combined with exit information, possible exit lanes can be obtained. Figure 1 As shown, there are 4 lanes at the exit, namely Left 1, Left 2, Left 3 and Left 4 from left to right. These lanes can all be used as target lanes for autonomous driving vehicles.

[0043] 8. When there is a difference in the number of lanes at the entrance and exit, the autonomous vehicle can choose from multiple lanes at the exit when passing through the intersection. Figure 1 In the example, depending on the entrance lane where the autonomous vehicle is located, it is feasible to choose the second left lane or the third left lane as the exit. Whether this choice is reasonable depends mainly on the driving mode of other vehicles traveling in the same direction as the vehicle. This situation can be Figure 2A and Figure 2B If only each vehicle is considered separately, the decision of the traffic mode is easily affected by the behavior of a single vehicle, which leads to unstable decision results. However, this method considers the vehicles in the same direction as a whole to obtain stable traffic flow information and input it into the decision system to provide more intelligent decision output. For example, for the same intersection, based on the overall driving conditions of the surrounding vehicles (such as Figure 2A and Figure 2B As shown in the figure, autonomous vehicles need to choose different exit lanes to reduce intersections with other vehicles, thereby improving the safety and efficiency of traffic at intersections.

[0044] 9. Based on the entrance and its corresponding lane information determined in steps 5 to 7 above, combined with the lane where the autonomous driving vehicle is located in step 2, the vehicle in front, the vehicle on the left, and the vehicle on the right can be determined. Figure 2A and Figure 2B In the figure, vehicle 3 (also marked with "left") is identified as the left vehicle. It should be noted that for vehicles 4 and 5 in the intersection, it is necessary to judge based on the historical lanes they were in before entering the intersection, and based on this, vehicles 4 and 5 can be identified as left vehicles; and vehicle 6 (also marked with "right") is the right vehicle. Similarly, for vehicles 7 and 8 in the intersection, based on the historical lanes they were in before entering the intersection, it can be determined that vehicles 7 and 8 are both right vehicles; vehicles 1 and 2 are both identified as front vehicles.

[0045] 10. Based on the current positions and historical track points of the front vehicle and the left and right vehicle groups distinguished in step 9, information about the vehicle group relative to the autonomous driving vehicle in the intersection can be obtained respectively. This information includes: the front vehicle (such as Figure 2A and Figure 2B Vehicles 1 and 2 in the left side (such as Figure 2A and Figure 2B Vehicles 3, 4, 5) and vehicles on the right (such as Figure 2A and Figure 2B Although the driving conditions of individual vehicles may vary, their overall driving trends are relatively stable.

[0046] a. Summarize the current coordinates and historical coordinates of the corresponding group of vehicles to obtain the overall traffic flow information of the corresponding group of vehicles. In this process, the historical coordinate set only covers the vehicles entering the threshold d mentioned in step 3. threshould The purpose of this is to prevent vehicle driving information that is not related to intersection traffic from interfering with the overall traffic trend.

[0047] b. Figure 3A and Figure 3B Given the Figure 2A and Figure 2B Examples of current and historical trajectory points of vehicle groups corresponding to the two modes of passage, where the left side shows the overall straight-ahead situation, and the right side shows the overall leftward situation. Each group of trajectory points is aggregated from the trajectories of multiple vehicles in the same group. The density of each group of trajectory points will vary depending on the perceived frame rate in actual applications, but in general, it can ensure that the distance between points is kept within 0.1m.

[0048] c. All the trajectory points here are the trajectories that the vehicle has actually traveled and do not contain any prediction information, so their information is deterministic.

[0049] 11. The corresponding grouped trajectory points obtained in step 10 are composed of the trajectory points of multiple vehicles. Although these trajectory points show a consistent trend as a whole, there may be some deviations in the details. In order to provide more stable trend information, an additional layer of information processing operation is required. One feasible method is to perform curve fitting on each set of trajectory points, so as to obtain a curve that can accurately reflect the driving trend.

[0050] Figure 4The overall driving trend obtained based on the trajectory points of different vehicle groups is shown. First, a local coordinate system is constructed with the autonomous vehicle as the center, and the vehicle trajectory points of different groups are converted to the local coordinate system. Then, a curve fitting problem is constructed in this local coordinate system, and the corresponding fitting curve is obtained by solving it. Here, the least squares fitting is introduced as an optional fitting method.

[0051] Assume that the trajectory of the vehicle roughly follows a polynomial curve of a certain order, which is assumed to be 4th order here. Then a point in the trajectory can be represented by the following expression: where x i ,y i is the coordinate value of the i-th trajectory point in the vehicle coordinate system, β j (j=0...4) is the fitting parameter. All trajectory points of the same group can be written in vector form Where X is a polynomial matrix of different orders composed of the x coordinates of all trajectory points, is the fitting parameter, and y is the column vector consisting of the y coordinates of all trajectory points.

[0052] By establishing the following objective function The coefficients of the polynomial can be found Next, the starting and ending values ​​of x are set according to the range of the intersection, and the corresponding step size is determined. The corresponding y value can then be calculated using a polynomial expression, so that the corresponding traffic flow fitting trend curve can be obtained.

[0053] 12. After completing step 11 and obtaining the fitted real-time trend curve of surrounding traffic, the decision module can intelligently select the mode of passage within the intersection based on the overall trend of surrounding vehicles and the location information of the exit lane of the intersection, which includes the distance of passage, feasible trajectory and the choice of exit lane. In this way, the possibility of potential interaction with vehicles passing in the same direction can be reduced, thereby improving the safety and efficiency of passage.

[0054] The overall process of the present invention is as follows Figure 5 shown.

[0055] Alternative implementation:

[0056] In the present invention, the information of vehicles in the same direction can be obtained by the travel direction of the historical lane, or by the similarity between the vehicle's driving posture and the self-vehicle.

[0057] In the present invention, when obtaining the overall driving trend curve according to the vehicle group trajectory point set, different fitting methods can be used. For example, a similar variant is to introduce a weight vector W according to the credibility of the surrounding vehicle trajectory points, and construct a weighted least squares problem: By solving this problem, when the weight vector W takes a reasonable value, a more reasonable trend curve can be obtained.

[0058] The use of the overall driving trend curve of the surrounding vehicle group in the present invention can also be extended to speed planning to guide the autonomous driving vehicle to maintain coordination with the surrounding traffic when passing through the intersection.

[0059] The beneficial technical effects brought by the technical solution of the present invention are:

[0060] (1) Improves traffic efficiency and avoids unnecessary confluence and interaction;

[0061] (2) Improved traffic safety, taking into account the overall driving patterns of surrounding vehicles and avoiding potential interaction risks;

[0062] (3) not relying on probabilistic forecast information;

[0063] (4) Increased the stability and reliability of intersection traffic decisions.

[0064] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An intelligent decision-making method for intersection traffic taking into account surrounding real-time traffic flow, characterized in that: The steps include: S1: Obtain status information of the autonomous vehicle and surrounding vehicles through sensors equipped by the autonomous vehicle; S2: Combine high-precision map information to place the vehicle and surrounding vehicles on a structured road, obtain lane information for the vehicle and other vehicles, and build lane-level relative relationships; S3: Obtain the distance from the vehicle to the intersection ahead based on the vehicle status and high-precision map information, and determine whether the distance is greater than a set threshold. If so, drive on a normal public road; otherwise, proceed to step S4 to estimate the surrounding real-time traffic flow; S4: Determine the required entrance and exit as well as the same-direction lane and exit lane information according to the traffic direction of the intersection ahead; S5: Filter out vehicles traveling in the same direction as the vehicle based on the real-time perception information and the lane information, and then distinguish the vehicle in front, the vehicle on the left, and the vehicle on the right based on the entrance and exit and the lane and exit lane information determined in step S4, and the lane information of the vehicle obtained in step S2; S6: obtaining information of a group of vehicles in front, a group of vehicles on the left, and a group of vehicles on the right relative to the vehicle according to the current positions and historical track points of the distinguished front vehicle, left vehicle, and right vehicle; S7: obtaining overall traffic flow information of the corresponding grouped vehicle group by summarizing the current and historical coordinates of the corresponding grouped vehicle group; S8: by performing curve fitting on the trajectory point set of each group of vehicles, a corresponding driving trend curve is obtained, thereby obtaining a driving trend curve of the surrounding traffic flow of the own vehicle; S9: Select the mode of passage within the intersection according to the driving trend curve of the surrounding traffic flow and the exit lane information.

2. The method according to claim 1, characterized in that In step S7, when the historical coordinates are summarized, only the coordinate information of the corresponding vehicle that enters the set threshold range mentioned in step S3 is included.

3. The method according to claim 1, characterized in that In step S8, when performing curve fitting on the trajectory point set, firstly, a local coordinate system is constructed with the vehicle as the center, and the trajectory point sets of different groups of vehicles are converted into the local coordinate system. Then, a curve fitting problem is constructed in this local coordinate system, and the corresponding fitting curve is obtained by solving it.

4. The method according to claim 3, characterized in that In step S8, assuming that the trajectory of the vehicle follows a 4th-order polynomial curve, a point in the trajectory is represented by the following expression: Among them, x i ,y i is the coordinate value of the i-th trajectory point in the vehicle coordinate system, β j (j=0...4) is the fitting parameter; First, all trajectory points of the same group of vehicles are written in vector form Among them, X is the polynomial matrix of different orders composed of the x coordinates of all trajectory points, For the parameters to be fitted, y is a column vector consisting of the y coordinates of all trajectory points, then the following objective function is established Find the coefficients of a polynomial Then, the starting and ending values ​​of x are set according to the range of the intersection, and the corresponding step size is determined. The corresponding y value is then calculated through a polynomial expression to obtain the corresponding fitting traffic flow trend curve.

5. The method according to claim 1, characterized in that In step S1, the sensors equipped by the autonomous driving vehicle include: lidar, high-definition camera, IMU / GNSS equipment; the status information of the vehicle includes: the position, velocity vector and acceleration vector of the vehicle; the status information of surrounding vehicles includes: the type, position, speed and relative relationship between other vehicles and the vehicle.

6. The method according to claim 1, characterized in that In step S2, the relative relationship between other vehicles and the vehicle includes: the front vehicle in the same lane; the rear vehicle in the same lane; the front vehicle in the right lane, the parallel vehicle in the right lane, and the rear vehicle in the right lane; the front vehicle in the left lane, the parallel vehicle in the left lane, and the rear vehicle in the left lane; Furthermore, when the road in the same direction has more lanes and there are other vehicles in the corresponding lanes, the relative relationship between the other vehicles and the vehicle is further expanded to cover all lanes in the same direction.

7. The method according to claim 6, characterized in that In step S2, if the roads in different directions in the intersection overlap, so that there are multiple roads in the intersection corresponding to the vehicle's position, resulting in unclear correspondence between other vehicles and the vehicle itself, it is necessary to trace back the historical lane information of other vehicles and mark them based on the relative relationship between the last non-intersection lane where the vehicle was before entering the intersection and the lane where the vehicle is located.

8. The method according to claim 1, characterized in that In step S9, the travel mode includes: travel distance, feasible trajectory and selection of exit lane.

9. An intelligent decision-making device for intersection traffic that takes into account surrounding real-time traffic flow, characterized in that: Includes the following modules: Status information acquisition module: obtains the status information of the autonomous vehicle and surrounding vehicles through the sensors equipped by the autonomous vehicle; Lane-level relative relationship building module: Combines high-precision map information to place the vehicle and surrounding vehicles on a structured road, obtains lane information for the vehicle and other vehicles, and builds lane-level relative relationships; Threshold judgment module: obtains the distance from the vehicle to the intersection ahead based on the vehicle status and high-precision map information, and determines whether the distance is greater than the set threshold. If so, the vehicle drives on normal public roads; otherwise, it needs to estimate the surrounding real-time traffic flow; Entrance and exit and lane determination module: When the threshold judgment module determines that real-time surrounding traffic flow estimation is required, the required entrance and exit as well as the same-direction lane and exit lane information are determined according to the traffic direction of the intersection ahead; Vehicle differentiation module: Based on real-time perception information and lane information, it selects vehicles traveling in the same direction as the vehicle, and then distinguishes the vehicle in front, the vehicle on the left, and the vehicle on the right according to the entrance and exit determined by the entrance and exit and lane determination module, as well as the lane and exit lane information of the vehicle in the same direction, and the lane information of the vehicle obtained in the lane-level relative relationship construction module; Vehicle group grouping module: based on the current positions and historical track points of the distinguished front vehicle, left vehicle and right vehicle, obtain the information of the front vehicle group, left vehicle group and right vehicle group relative to the vehicle; Traffic flow information acquisition module: by summarizing the current and historical coordinates of the corresponding grouped vehicle groups, the overall traffic flow information of the corresponding grouped vehicle groups is obtained; Driving trend curve fitting module: by performing curve fitting on the trajectory point set of each group of vehicles, the corresponding driving trend curve is obtained, thereby obtaining the driving trend curve of the surrounding traffic of the vehicle; Traffic mode selection module: selects the traffic mode within the intersection based on the driving trend curve of the surrounding traffic flow and the exit lane information.

10. A computer-readable storage medium containing a computer program, characterized in that: When the computer program is executed by one or more processors, the intelligent decision-making method for intersection traffic taking into account surrounding real-time traffic flow as described in any one of claims 1 to 8 is implemented.

Images