CN115583258A

CN115583258A - Automatic vehicle meeting control method and device, vehicle control equipment and medium

Info

Publication number: CN115583258A
Application number: CN202211393685.1A
Authority: CN
Inventors: 林腾波; 卢天明
Original assignee: Guangzhou Weride Technology Co Ltd
Current assignee: Guangzhou Weride Technology Co Ltd
Priority date: 2022-11-08
Filing date: 2022-11-08
Publication date: 2023-01-10

Abstract

The invention discloses a method and a device for controlling meeting of an automatic driving vehicle, vehicle control equipment and a medium, wherein the method comprises the following steps: when the situation that a target vehicle exists at an intersection in front of a main vehicle and on the side of the main vehicle lane far away from an opposite lane is detected, a first predicted track and a second predicted track of the target vehicle running from the intersection to the main vehicle lane and the opposite lane are respectively generated, a first similarity of the running track of the target vehicle and the first predicted track is calculated, a second similarity of the running track of the target vehicle and the second predicted track is calculated, the running intention of the target vehicle is determined according to the first similarity and the second similarity, and the problem that collision risk exists when sudden braking of the automatic driving vehicle is caused because the automatic driving vehicle cannot predict the running intention of the target vehicle running from the intersection is solved according to control of the running intention, so that the automatic driving vehicle decelerates and avoids the sudden braking when the running intention of the target vehicle is predicted, collision is avoided, and the safety of the automatic driving vehicle meeting at the main vehicle is improved.

Description

Automatic vehicle meeting control method and device, vehicle control equipment and medium

Technical Field

The invention relates to the technical field of automatic driving vehicles, in particular to a meeting control method and device for an automatic driving vehicle, vehicle control equipment and a medium.

Background

As more and more autonomous vehicles are available, the autonomous vehicles inevitably meet other social vehicles at intersections during the course of traveling on roads.

As shown in fig. 1, there is an intersection R3 on the right side of a host vehicle lane R1 where a host vehicle a (autonomous vehicle) is located, the host vehicle a cannot accurately recognize whether the travel intention of a target vehicle B at the intersection R3 is to travel from the intersection left to the opposite lane R2 or from the intersection right to the host vehicle lane R1, and if the host vehicle a cannot recognize the travel intention of the target vehicle B, the host vehicle a cannot avoid the target vehicle B that is coming from the intersection R3 in advance, causing sudden braking before the host vehicle a travels to the intersection R3, and there is a risk of collision with the target vehicle B.

Disclosure of Invention

The invention provides a method and a device for controlling meeting of an automatic driving vehicle, vehicle control equipment and a medium, which are used for solving the problem that in the prior art, when the automatic driving vehicle runs to an intersection, the driving intention of the vehicle running out of the intersection cannot be predicted, so that the automatic driving vehicle is subjected to sudden braking and has collision risk.

In a first aspect, the present invention provides a method for controlling vehicle crossing of an autonomous vehicle, comprising:

when a side of a main vehicle lane far from an opposite lane and a target vehicle at an intersection in front of the main vehicle are detected, respectively generating a first predicted track and a second predicted track of the target vehicle driving from the intersection to the main vehicle lane and the opposite lane;

acquiring a running track of the target vehicle;

calculating a first similarity of the travel track and the first predicted track, and calculating a second similarity of the travel track and the second predicted track;

determining the driving intention of the target vehicle according to the first similarity and the second similarity, and controlling the host vehicle according to the driving intention.

In a second aspect, the present invention provides an automatic vehicle-crossing control device, comprising:

the predicted track generation module is used for respectively generating a first predicted track and a second predicted track of the target vehicle running from the intersection to the main vehicle lane and the opposite lane when the side, far away from the opposite lane, of the main vehicle lane and the target vehicle at the intersection in front of the main vehicle are detected;

the driving track acquisition module is used for acquiring the driving track of the target vehicle;

the similarity calculation module is used for calculating a first similarity between the running track and the first predicted track and calculating a second similarity between the running track and the second predicted track;

and the driving control module is used for determining the driving intention of the target vehicle according to the first similarity and the second similarity and controlling the main vehicle according to the driving intention.

In a third aspect, the invention provides a vehicle control apparatus including:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein, the first and the second end of the pipe are connected with each other,

the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of controlling an automated vehicle crossing according to the first aspect of the present invention.

In a fourth aspect, the present invention provides a computer readable storage medium storing computer instructions for causing a processor to implement the method for controlling crossing of autonomous vehicles according to the first aspect of the present invention when executed.

According to the embodiment of the invention, when the side of the main vehicle lane far from the opposite lane and the intersection in front of the main vehicle are detected to have the target vehicle, the first predicted track and the second predicted track of the target vehicle running from the intersection to the main vehicle lane and the opposite lane are respectively generated, the first similarity between the running track of the target vehicle and the first predicted track is calculated, the second similarity between the running track of the target vehicle and the second predicted track is calculated, the running intention of the target vehicle is determined according to the first similarity and the second similarity, and the main vehicle is controlled according to the running intention.

It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present invention, nor do they necessarily limit the scope of the invention. Other features of the present invention will become apparent from the following description.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a schematic diagram of a vehicle meeting scene at an intersection according to an embodiment of the present invention;

fig. 2 is a flowchart of an automatic vehicle meeting control method according to an embodiment of the present invention;

fig. 3A is a flowchart of an automatic vehicle meeting control method according to a second embodiment of the present invention;

FIG. 3B is a schematic diagram of intersection detection;

FIG. 3C is a schematic illustration of the calculated similarity of the driving trajectory and the predicted trajectory in an embodiment of the present invention;

FIG. 3D is a schematic illustration of a travel path for avoidance of a target vehicle in an embodiment of the present invention;

fig. 4 is a schematic structural diagram of an automatic vehicle meeting control device according to a third embodiment of the present invention;

fig. 5 is a schematic configuration diagram of a vehicle control apparatus provided in a fourth embodiment of the present invention.

Detailed Description

In order to make those skilled in the art better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Example one

Fig. 2 is a flowchart of an automatic vehicle-meeting control method according to an embodiment of the present invention, where the automatic vehicle-meeting control method is applicable to controlling a situation that an automatic vehicle meets a vehicle running out of an intersection, and the method can be executed by an automatic vehicle-meeting control device, where the automatic vehicle-meeting control device can be implemented in hardware and/or software, and the automatic vehicle-meeting control device can be configured in a vehicle control device, and the vehicle control device can be a driving computer of the automatic vehicle, or a remote server communicating with the automatic vehicle. As shown in fig. 2, the automatic vehicle meeting control method includes:

s201, when the side, far away from the opposite lane, of the main lane and the existence of a target vehicle at an intersection in front of the main vehicle are detected, a first predicted track and a second predicted track of the target vehicle, which travel from the intersection to the main lane and the opposite lane, are respectively generated.

The main vehicle is an automatic driving vehicle, the lane of the main vehicle is a lane on which the main vehicle runs, the opposite lane is a lane adjacent to the lane of the main vehicle and the running direction of the opposite lane is opposite to that of the lane of the main vehicle, for the rule of running on the right, the side of the lane of the main vehicle far away from the opposite lane is the right side of the lane of the main vehicle, for the rule of running on the left, the side of the lane of the main vehicle far away from the opposite lane is the left side of the lane of the main vehicle, and the right running rule is taken as an example in the embodiment.

As shown in fig. 1, a host vehicle a traveling on a host vehicle lane R1 can detect whether there is an intersection in front of the host vehicle and on the right side of the host vehicle lane R1, as shown in fig. 1, can detect whether there is an intersection R3 in front of the host vehicle a and detect whether there is a target vehicle B whose head travels toward the host vehicle lane R1 at the intersection R3, if there is a target vehicle B, a first predicted trajectory L1 is generated in which the target vehicle B travels from the intersection R3 to the host vehicle lane R1, and a second predicted trajectory L2 is generated in which the target vehicle B travels from the intersection R3 to an oncoming lane R2, that is, predicted travel trajectories in which the target vehicle B turns right and left from the intersection R3.

In an alternative embodiment, the host vehicle a may sense the environment through sensors to obtain environment data, and recognize from the environment data whether there is a target vehicle B having a head directed toward the host vehicle lane R1 at the intersection R3, and when there is a target vehicle B at the intersection R3, acquire data of the position, speed, acceleration, head direction, and the like of the target vehicle to generate a first predicted trajectory L1 for the target vehicle B to turn right to travel from the intersection R3 to the host vehicle lane R1 and to generate a second predicted trajectory L2 for the target vehicle B to turn left to travel from the intersection R3 to the oncoming lane.

In one example, the host vehicle a may determine, through the environment data, a position after the target vehicle B has traveled right to exit the intersection R3 as an end point P1 of the first predicted trajectory, and a position after the target vehicle B has traveled left to exit the intersection to reach the opposing lane R3 as an end point P2 of the second predicted trajectory, and then after acquiring the state data of the target vehicle B, plan the first predicted trajectory L1 and the second predicted trajectory L2 in conjunction with the environment data through a trajectory planning algorithm such as a search algorithm, a quadratic optimization algorithm, or the like, and update the first predicted trajectory L1 and the second predicted trajectory L2 in real time.

S202, acquiring the running track of the target vehicle.

The running track of the target vehicle is the actual running track of the target vehicle within a preset time length, the preset time length can be equal to the time length of the updating period of the first predicted track and the second predicted track, and the main vehicle can acquire data such as the position, the speed, the acceleration, the head orientation and the like of the target vehicle at multiple moments through a sensor to generate the running track.

S203, calculating a first similarity of the running track and the first predicted track, and calculating a second similarity of the running track and the second predicted track.

The driving track is an actual driving track of a target vehicle within a preset time length, the driving track has a starting time and an ending time, the track from the starting time to the ending time can be intercepted from the first predicted track, the distance of state data such as the position, the speed, the acceleration, the vehicle head orientation and the like of track points of the intercepted track and the driving track at the same time is calculated to serve as the similarity of the two track points, the sum of the similarities of the plurality of track points is calculated to serve as a first similarity of the driving track and the first predicted track, and similarly, a second similarity of the driving track and a second predicted track is calculated.

In another example, in calculating the first similarity, after calculating a sum of the similarities of the multiple track points, an indication value corresponding to the state of the turn signal of the target vehicle may be further obtained, and after a weight is given to the sum of the indication value and the similarities of the multiple track points, the second similarity may be calculated by calculating a weighted sum of the sum and the indication value as the first similarity.

And S204, determining the driving intention of the target vehicle according to the first similarity and the second similarity, and controlling the host vehicle according to the driving intention.

The first similarity represents the probability that the target vehicle runs according to the first predicted track, namely the probability that the target vehicle runs straight from the intersection to the host lane, the second similarity represents the probability that the target vehicle runs according to the second predicted track, namely the probability that the target vehicle runs from the intersection to the target lane, whether the first similarity and the second similarity are larger than a preset threshold value or not can be judged respectively, when the first similarity is larger than the preset threshold value, the target vehicle is supposed to run in the host lane after turning right from the intersection, whether the target vehicle needs to wait to run to the host lane or not can be determined according to the original running track of the host vehicle and the running track of the target vehicle, and after the target vehicle runs in the same direction with the host vehicle, the host vehicle is controlled to continue running according to the original running track or the target vehicle is avoided by means of the opposite lane. When the second similarity is larger than the preset threshold value, the target vehicle is shown to run from the left turn of the intersection to the opposite lane, and when the running track of the main vehicle and the track of the target vehicle turning to enter the opposite lane have an overlapping area, the main vehicle can be controlled to reduce the target vehicle giving way in advance, so that the risk of collision caused by sudden braking when the main vehicle meets the target vehicle is avoided.

Example two

Fig. 3A is a flowchart of an automated vehicle meeting control method according to a second embodiment of the present invention, which is optimized based on the first embodiment of the present invention, and as shown in fig. 3A, the automated vehicle meeting control method includes:

s301, in the process of driving of the host vehicle, detecting whether an intersection exists on one side of the host vehicle lane far away from the opposite lane and in front of the host vehicle.

In an optional embodiment of the invention, a lane center line of a main vehicle lane within a preset range in front of the main vehicle and a road edge of the main vehicle lane far away from an opposite lane can be acquired, the lane center line is sampled to obtain a plurality of sampling points, the distance from each sampling point to the road edge is calculated, and when the distance is greater than a preset distance threshold value, it is determined that an intersection exists on one side of the main vehicle lane far away from the opposite lane and in front of the main vehicle.

As shown in fig. 3B, a lane center line X1 of the host lane R1 and a road edge X3 of the host lane R1 may be obtained from the road network map, and a distance d from a point on the lane center line X1 in front of the host vehicle a to the road edge X3 is calculated, and when d is greater than a set value, it indicates that the intersection R3 is detected, and the intersection is detected by calculating the distance from the lane center line to the road edge, which is suitable for roads with relatively perfect roads and road edges.

In another alternative embodiment, the lane central line of the main vehicle lane can be found in a preset semantic map, the current position and the driving direction of the main vehicle are mapped into the semantic map, and whether other lane central lines which are far away from one side of the opposite lane and connected with the lane central line exist in the preset range of the lane central line in the driving direction of the current position or not is judged; if yes, determining that the side of the main vehicle lane far away from the opposite lane and the intersection in front of the main vehicle exist, and identifying the intersection through the information on the semantic map, wherein the data volume is small, and the accuracy and the efficiency are high.

As shown in fig. 3B, the lane centerlines X2, X4, and X5 are connected to the lane centerline X1 of the host lane R1 in front of the host vehicle a, and it can be determined that an intersection is detected, where the lane centerline X2 is used for right-turn entry into the intersection R3 on the host lane R1, the lane centerline X4 is used for the vehicle at the intersection R3 to travel from the intersection R3 to the host lane R1, and the lane centerline X5 is used for the vehicle at the intersection R3 to travel from the intersection R3 to the opposite lane R2.

In yet another alternative embodiment, the intersection may be identified by inputting environmental data sensed by a sensor on the host vehicle into the intersection identification model, and a person skilled in the art may also identify the intersection in other ways, which is not limited in the embodiment of the present invention.

And S302 is executed when the side of the main vehicle lane far away from the opposite lane and the intersection in front of the main vehicle are detected, and the main vehicle is controlled to continue to run according to the original running track when the side of the main vehicle lane far away from the opposite lane and the intersection in front of the main vehicle are detected.

S302, detecting whether a target vehicle with a vehicle head running towards the main vehicle lane exists at the intersection.

In an alternative embodiment, the host vehicle may sense the environment including the intersection through a sensor, obtain environment data, such as point cloud data, input the point cloud data into the vehicle recognition model to recognize whether a vehicle is included at the intersection, and recognize the head orientation of the vehicle, when it is recognized that there is a target vehicle at the intersection, which is driven by the head towards the host vehicle lane, and it is required to predict whether the target vehicle turns left or right, S303 is performed, and if no target vehicle is detected, the host vehicle is controlled to drive according to the original track.

And S303, respectively generating a first predicted track and a second predicted track of the target vehicle driving from the intersection to the main lane and the opposite lane.

Alternatively, a first target position and a second target position may be determined, the first target position being a position after the target vehicle travels from the intersection to the host lane, the second target position being a position after the target vehicle travels from the intersection to the oncoming lane, the first target position and the second target position being end positions, respectively, with the current position of the target vehicle being a starting position, and the current state of the target vehicle being an input of the trajectory planning algorithm, to generate a first predicted trajectory and a second predicted trajectory, respectively, for the target vehicle to travel from the intersection to the host lane and the oncoming lane.

As shown in fig. 1, the current position of the target vehicle B at the intersection R3 is P0, the first target position is P1, and the second target position is P2, it is possible to generate a first predicted trajectory L1 for the target vehicle B to exit the intersection R3 for a right turn into the host lane R1 with the first target position P1 as the end position, with the current position P0 as the start position, and to generate a second predicted trajectory L2 for the target vehicle B to exit the intersection R3 for a left turn into the opposite lane R2 with the second target position P2 as the end position, with the current position P0 as the start position. In one example, the first predicted trajectory L1 and the second predicted trajectory L2 may be generated by a path planning algorithm such as a search algorithm, a quadratic optimization algorithm, and the like, using data such as a current position, a velocity, an acceleration, and the like of the target vehicle B as state data, establishing constraints such as an outer dimension of the target vehicle B, an obstacle perceived by the host vehicle, and the like.

It should be noted that the first predicted trajectory L1 and the second predicted trajectory L2 may be updated according to the state data of the target vehicle B at a preset cycle.

And S304, acquiring the running track of the target vehicle.

The target vehicle is a real target vehicle running track within a preset time length, the preset time length can be equal to the time length of the updating period of the first predicted track and the second predicted track, and the main vehicle can acquire data of the position, the speed, the acceleration, the head orientation and the like of the target vehicle at a plurality of moments through a sensor to generate the running track.

And S305, acquiring the starting time and the ending time of the running track.

In this embodiment, the travel track of the target vehicle is an actual travel track within a preset time period, the travel track has a start time and an end time, and as shown in fig. 3C, the travel track is a schematic diagram of a first predicted track L1, a second predicted track L2 and a travel track L3, the start time of the travel track L3 is t1, and the end time of the travel track L3 is t2.

S306, respectively intercepting the track from the starting time to the ending time from the first predicted track and the second predicted track to obtain a first sub-track and a second sub-track.

As shown in FIG. 3C, a first sub-track m1-m2 is obtained by cutting the track from the start time t1 to the end time t2 from the first predicted track L1, and a track n1-n2 is obtained by cutting the track from the start time t1 to the end time t2 from the second predicted track L2.

S307, calculating the similarity between the driving track and the first sub-track to obtain a first similarity, and calculating the similarity between the driving track and the second sub-track to obtain a second similarity.

Taking the calculation of the similarity between the running track and the first sub-track to obtain the first similarity as an example, acquiring an indicated value corresponding to the state of a steering lamp on one side of a target vehicle from a starting time to an ending time, respectively sampling the running track and the first sub-track according to a preset time interval to obtain a plurality of running track points on the running track and sub-track points on the first sub-track, wherein each running track point and each sub-track point contain at least one item of state data, and then calculating the similarity between the running track points and the sub-track points at the same moment by the following formula (1):

in the above formula (1), D _i For the similarity of the driving track point i and the sub-track point i at the same moment, N represents N items of state data in total, A _j Weight, X, representing the jth item of status data _1ij Item j of state data, X, representing a point i of the track _2ij And j-th item of state data representing the sub track point i.

After the similarity of the driving track and the sub-track at the same moment is calculated, the sum of the similarities of the driving track points and the sub-track points at a plurality of sampling moments is calculated, and the similarity of the driving track and the first sub-track is calculated through the following formula (2) to obtain a first similarity:

in the above formula (2), S is a first similarity, a is a weight of a sum of similarities of a plurality of trace points calculated from state data at k times, C is an indication value of a turn signal state, C is smaller than 0 when the turn signal is blinking, C is larger than 0 when the turn signal is not blinking, and b is a weight of the turn signal state.

For example, as shown in fig. 3C, assuming that there are k track points on the travel track L3 and k track points on the first sub-tracks m1-m2, there are j pieces of state data for the travel track point and the sub-track point at the same time i, and the j pieces of state data include at least one of speed, acceleration, displacement, position, head orientation, and the like, then for the travel track point and the sub-track point sampled at each time, the above formula (1) can be appliedCalculating pixel point D of track point pair sampled at the moment _i Then, as for the similarity of the running locus L3 and the first sub-locus m1-m2, which can be calculated by the above-mentioned formula (2), as shown in fig. 1 and 3C, for the host vehicle a, the host vehicle a detects the left winker of the target vehicle B more accurately, and when the state of the left winker is blinking, which indicates that the target vehicle B is more likely to turn left into the opposing lane R2, the first similarity is accordingly decreased, that is, the indicated value C indicating the state of the winker is given less than 0, such as-0.5, so that the running locus L3 is less likely to turn left than the first sub-locus m1-m2, whereas when the state of the left winker is not blinking, which indicates that the target vehicle B is more likely to turn right into the host vehicle lane R1, the first similarity is correspondingly increased, that the indicated value C given the state of the winker is greater than 0, such as +0.5.

Similarly, in calculating the second similarity, when the left turn signal state is blinking, indicating that the target vehicle B is more likely to turn left into the opposing lane R2, the second similarity is increased accordingly, that is, the indication value C indicating that the turn signal state can be given is greater than 0, such as +0.5, so that the traveling locus L3 is more likely to be similar to the second sub-trajectories n1-n2, whereas, when the left turn signal state is not blinking, indicating that the target vehicle B is more likely to turn right into the main lane R1, the second similarity is decreased accordingly, that is, the indication value C indicating that the turn signal state can be given is less than 0, such as-0.5.

According to the method and the device, the similarity of the track is calculated through the multiple dimension state data such as the acceleration, the speed, the displacement, the coordinate and the direction of the vehicle head of the track point, the driving intention of the target vehicle can be reflected from the multiple dimension state data, the similarity can accurately reflect the driving intention of the target vehicle, and the accuracy of the prediction of the driving intention of the target vehicle by the main vehicle is improved.

Of course, in another example, after the driving trajectory L3 and the first sub-trajectories m1 to m2 are patterned, the similarity of the images may be calculated as the first similarity of the driving trajectory L3 and the first sub-trajectories m1 to m2, and the embodiment does not limit the way of calculating the similarity.

S308, judging whether the first similarity and the second similarity are larger than a similarity threshold value.

Specifically, it may be determined whether or not the first similarity and the second similarity are greater than the similarity threshold value, respectively, and in this embodiment, the travel intention of the target vehicle is to travel from the intersection right into the host lane, or from the intersection left into the opposite lane, that is, as shown in fig. 1, the target vehicle finally travels along one of the first predicted trajectory L1 and the second predicted trajectory L2, and when one of the first similarity and the second similarity is larger, the other is smaller, that is, when one of the first similarity and the second similarity is greater than the similarity threshold value, the other is smaller than the similarity threshold value, S309 and S310 are performed when the first similarity is greater than the similarity threshold value, S311 and S312 are performed when the second similarity is greater than the similarity threshold value, and when both the first similarity and the second similarity are less than the similarity threshold value, the travel intention of the target vehicle cannot be decided, and the host vehicle may be controlled to wait for deceleration until the travel intention of the target vehicle is decided.

S309, when the first similarity is larger than the similarity threshold value, determining that the target vehicle is driven to the main lane from the intersection.

If the first similarity is larger than the similarity threshold value, the data of the acceleration, the speed, the displacement, the coordinates, the head orientation and the like of the target vehicle are closer to the acceleration, the speed, the displacement, the coordinates and the head orientation of the corresponding track point in the first predicted track, and the driving intention of the target vehicle can be determined to be that the target vehicle drives in the same direction as the host vehicle after turning right from the intersection to the host vehicle lane.

And S310, controlling the running of the host vehicle according to the first predicted track and the original track of the host vehicle, wherein the original track is the running track of the host vehicle when the target vehicle is not detected.

In an optional embodiment, when the overlapping running area of the main vehicle and the oncoming vehicle is determined according to the first predicted track and the original track, detecting whether the oncoming vehicle exists in the oncoming lane, if so, controlling the main vehicle to decelerate, and driving through an intersection along with the target vehicle after the target vehicle drives to the main vehicle lane; if not, planning a target running track, wherein the target running track is a track passing through the opposite lane to avoid the target vehicle and then returning to the main lane, and controlling the main vehicle to run according to the target running track.

As shown in fig. 3D, the first predicted trajectory is L1, the original trajectory is L4, and as can be seen from the entering 3D, the original trajectory L4 of the host vehicle a and the first predicted trajectory L1 of the target vehicle B spatially and temporally have an overlapping travel area D, and if the host vehicle a travels along the original trajectory L4, the host vehicle a may collide with the target vehicle B, in which case if there is an oncoming vehicle in the oncoming lane R2, the host vehicle a may be controlled to stop traveling before reaching the overlapping travel area D, and after waiting for the target vehicle B to travel to enter the overlapping travel area D into the host lane R1 and travel in the same direction as the host vehicle a, the host vehicle a may be controlled to travel through the road junction R3 following the target vehicle B, so as to avoid the host vehicle a from traveling in the way with the target vehicle B, and to improve the safety of traveling of the host vehicle a.

If there is no oncoming vehicle in the oncoming lane R2, the host vehicle a can travel to the oncoming lane R2 to avoid the overlapping travel area D, and after avoiding the overlapping travel area D, the host vehicle a can travel from the oncoming lane R2 back to the host vehicle lane R1, as shown in fig. 3D, and the host vehicle a can replan the trajectory to obtain a trajectory L5, control the host vehicle a to travel according to the trajectory L5, without waiting for the target vehicle B, and improve the traveling efficiency of the host vehicle a.

And S311, when the second similarity is larger than the similarity threshold value, determining that the target vehicle is driven from the intersection to the opposite lane.

If the second similarity is larger than the similarity threshold value, the acceleration, speed, displacement, coordinates, head orientation and other data of the target vehicle are closer to the acceleration, speed, displacement, coordinates and head orientation of the corresponding track point in the second predicted track, and the driving intention of the target vehicle can be determined to be driving from the intersection left turn to the opposite lane.

And S312, controlling the main vehicle to stop running so as to avoid the target vehicle.

The target vehicle turns left to enter the opposite lane, so that the target vehicle can cross the whole main vehicle lane, the intersection overlapping running area with the main vehicle is large, the safety of the main vehicle avoiding is low, the main vehicle can be controlled to stop running before reaching the intersection so as to avoid the target vehicle, and after the target vehicle runs to the opposite lane, the main vehicle is controlled to continue running according to the original track so as to improve the running safety of the main vehicle.

Of course, the passing cost of the main vehicle for avoiding the target vehicle can also be calculated, the passing cost of the target vehicle for avoiding the main vehicle is calculated, and the party with low passing cost avoids the party with high passing cost.

When a target vehicle with a vehicle head running towards a main vehicle lane is detected at an intersection in front of the main vehicle and on one side far away from an opposite lane of the main vehicle lane, a first predicted track and a second predicted track of the target vehicle running from the intersection to the main vehicle lane and the opposite lane are respectively generated, the first sub-track is intercepted at the first predicted track, the second sub-track is intercepted at the second predicted track, the similarity between the first sub-track and the running track and the similarity between the second sub-track and the running track are respectively calculated to obtain a first similarity and a second similarity, the running intention of the target vehicle is determined according to the first similarity and the second similarity, the main vehicle is controlled according to the running intention, the problem that the running intention of the target vehicle running from the intersection cannot be predicted by an automatic driving vehicle is solved, the collision risk of the automatic driving vehicle exists, the automatic driving vehicle is decelerated and avoided when the running intention of the target vehicle running from the intersection is predicted, the collision of the automatic driving vehicle is avoided, and the safety of the automatic driving vehicle at the intersection and the meeting comfort are improved.

Further, when the first similarity is larger than the similarity threshold value, when an overlapped running area of the main vehicle and the opposite vehicle is determined according to the first predicted track and the original track, whether the opposite vehicle exists in the opposite lane is detected, if yes, the main vehicle is controlled to decelerate, the main vehicle runs through the intersection along with the target vehicle after running to the main vehicle lane, the main vehicle and the target vehicle are prevented from rushing to run, the running safety of the main vehicle is improved, if not, the target running track is planned, the target running track is a track which passes through the opposite lane to avoid the target vehicle and then returns to the main vehicle lane, the main vehicle is controlled to run according to the target running track, the target vehicle does not need to wait to run through the intersection, and the running efficiency of the main vehicle is improved.

Furthermore, when the similarity is calculated, different indicated values are given according to the states of the steering lamps of the target vehicle to calculate the similarity, so that the calculated similarity can more truly and accurately represent the driving intention of the target vehicle, the accuracy of the main vehicle in making a decision according to the similarity is improved, and the safety of vehicle crossing between the main vehicle and the target vehicle is improved.

In another alternative embodiment, the host vehicle may acquire state data of the target vehicle within a preset time period when detecting that the target vehicle exists at an intersection in front of the host vehicle on a side of the host vehicle lane away from the opposite lane, input the state data into a pre-trained travel intention recognition model to obtain a travel intention of the target vehicle, generate a third predicted trajectory of the target vehicle based on the travel intention, and control the host vehicle according to the third predicted trajectory.

Wherein the vehicle travel intention recognition model may be a model that outputs the travel intention of the target vehicle after inputting the state data of the target vehicle, and particularly in the present embodiment, the vehicle travel intention model may be a model that recognizes the travel intention of the target vehicle as a left turn or a right turn, and in practical applications, state data of a plurality of vehicles that are traveling right turns into the host lane from the intersection and left turns into the opposite lane may be obtained, the state data may be data of the speed, acceleration, position, steering wheel angle, turn signal state, etc. of the vehicle for a preset period of time, and the state data of each vehicle is labeled with the travel intention, for example, the travel intention of the state data of the vehicle traveling right turns into the host lane from the intersection is right turns into the host lane, the method comprises the steps that the driving intention of state data of a vehicle which drives out of an intersection and turns right to enter a main vehicle lane is a left-turn entering opposite lane, when a vehicle driving intention model is trained, the state data of the vehicle is input into the vehicle driving intention model to predict the driving intention of the vehicle, errors are calculated according to the predicted driving intention and the real driving intention of the vehicle, model parameters are further adjusted through the errors until the accuracy of the model for predicting the driving intention of the vehicle is larger than a preset threshold value, a trained vehicle driving intention recognition model is obtained, after state data of a target vehicle located at the intersection for a preset time period are obtained, the state data can be input into the trained vehicle driving intention recognition model, and the target vehicle driving intention is obtained and is the right-turn entering main vehicle lane or the left-turn entering opposite lane.

After determining the travel intention of the target vehicle, a third predicted trajectory may be generated, and in one example, as shown in fig. 1, when the travel intention is to turn right into the host lane, the third predicted trajectory is a trajectory L1 that turns right from the current position P0 of the intersection R3 into the position P1 of the host lane R1, and when the travel intention is to turn left into the oncoming lane R2, the third predicted trajectory is a trajectory L2 that turns left from the current position P0 of the intersection R3 into the position P2 of the oncoming lane R2, so that the travel of the host vehicle a may be controlled according to the trajectory L1 or L2, and the specific control process may refer to S301 to S312.

According to the embodiment, the driving intention of the target vehicle can be directly recognized in the driving intention recognition model through the state data input by the target vehicle within the preset time length, the main vehicle is controlled after the predicted track is generated through the driving intention, namely the driving intention is directly obtained through the state data input model to generate the predicted track, the predicted track is used for controlling the main vehicle to stop driving and avoiding or avoid a lane by-pass, the data processing amount is small, and the efficiency is high.

Example four

Fig. 4 is a schematic structural diagram of an automatic vehicle meeting control device according to a fourth embodiment of the present invention. As shown in fig. 4, the automatic vehicle-crossing control device includes:

a predicted trajectory generation module 401, configured to generate a first predicted trajectory and a second predicted trajectory for a target vehicle to travel from an intersection to a host lane and a oncoming lane, respectively, when it is detected that the host lane is on a side away from the oncoming lane and there is the target vehicle at the intersection in front of the host vehicle;

a driving track obtaining module 402, configured to obtain a driving track of the target vehicle;

a similarity calculation module 403, configured to calculate a first similarity between the driving trajectory and the first predicted trajectory, and calculate a second similarity between the driving trajectory and the second predicted trajectory;

a driving control module 404, configured to determine a driving intention of the target vehicle according to the first similarity and the second similarity, and control the host vehicle according to the driving intention.

Optionally, the method further comprises:

the intersection detection module is used for detecting whether an intersection exists on one side of the main vehicle lane far away from the opposite lane and in front of the main vehicle in the driving process of the main vehicle;

and a target vehicle detection module, configured to detect whether there is a target vehicle at the intersection that has a vehicle head that travels towards the host vehicle lane, and execute the predicted trajectory generation module 401 when there is a target vehicle at the intersection that has a vehicle head that travels towards the host vehicle lane.

Optionally, the predicted trajectory generating module 401 includes:

a target position determination unit for determining a first target position and a second target position, the first target position being a position after the target vehicle travels from the intersection to the host lane, the second target position being a position after the target vehicle travels from the intersection to the oncoming lane;

a predicted trajectory generating unit, configured to generate a first predicted trajectory and a second predicted trajectory of the target vehicle traveling from the intersection to the host lane and the opposite lane, respectively, using a current position of the target vehicle as a start position, the first target position and the second target position as end positions, respectively, and a current state of the target vehicle as an input of a trajectory planning algorithm.

Optionally, the driving track is a track acquired according to a preset period, and the similarity calculation module 403 includes:

the starting time and ending time obtaining submodule is used for obtaining the starting time and the ending time of the running track;

the sub-track intercepting submodule is used for respectively intercepting tracks from the starting time to the ending time from the first predicted track and the second predicted track to obtain a first sub-track and a second sub-track;

and the similarity operator module is used for calculating the similarity between the driving track and the first sub-track to obtain a first similarity, and calculating the similarity between the driving track and the second sub-track to obtain a second similarity.

Optionally, the similarity operator module comprises:

a turn signal indicating value obtaining unit, configured to obtain an indicating value corresponding to a turn signal state of a turn signal of the target vehicle on a side close to the host vehicle from the start time to the end time;

the track point sampling unit is used for respectively sampling the traveling track and the first sub-track according to a preset time interval to obtain a plurality of traveling track points on the traveling track and sub-track points on the first sub-track, and each traveling track point and each sub-track point comprise at least one item of state data;

the track point similarity calculation unit is used for calculating the similarity of the driving track point and the sub-track point at the same moment according to the following formula:

in the above formula, D _i For the similarity of the driving track point i and the sub-track point i at the same moment, N represents N items of state data in total, A _j Weight, X, representing the jth item of status data _1ij Item j of state data, X, representing a point i of the track _2ij J-th item of state data representing the sub track point i;

the similarity sum value calculating unit is used for calculating the sum value of the similarity of the driving track point and the sub-track points at a plurality of sampling moments;

a first similarity calculation unit, configured to calculate a similarity between the travel track and the first sub-track to obtain a first similarity by using the following formula:

in the above formula, S is a first similarity, a is a weight of a sum of similarities of a plurality of trace points calculated from state data at k times, C is an indication value of a turn signal state, C is smaller than 0 when the turn signal is blinking, C is larger than 0 when the turn signal is not blinking, and b is a weight of the turn signal state.

Optionally, the driving control module 404 includes:

the similarity judging submodule is used for judging whether the first similarity and the second similarity are greater than a similarity threshold value or not;

a first travel intention determination submodule for determining that the travel intention of the target vehicle is to travel from the intersection to the host lane when the first similarity is greater than the similarity threshold;

a first driving control sub-module, configured to control driving of the host vehicle according to the first predicted trajectory and an original trajectory of the host vehicle, where the original trajectory is a driving trajectory of the host vehicle when the target vehicle is not detected;

a second driving intention determining submodule for determining that the driving intention of the target vehicle is to drive from the intersection to the opposite lane when the second similarity is greater than the similarity threshold;

and the second running control submodule is used for controlling the main vehicle to stop running so as to avoid the target vehicle.

Optionally, the first driving control sub-module comprises:

a oncoming vehicle detection unit configured to detect whether there is a oncoming vehicle in the oncoming lane, when it is determined that there is an overlapping travel area between the host vehicle and the oncoming vehicle from the first predicted trajectory and the original trajectory;

a first control unit for controlling the host vehicle to decelerate and to follow the target vehicle to travel through the intersection after the target vehicle travels to the host vehicle lane;

a driving track planning unit, configured to plan a target driving track, where the target driving track is a track that passes through the opposite lane to avoid the target vehicle and returns to the main lane;

and the second control unit is used for controlling the main vehicle to run according to the target running track.

Optionally, the method further comprises:

the state data acquisition module is used for acquiring state data of a target vehicle within a preset time length when the condition that the target vehicle exists at an intersection in front of a main vehicle on one side of the main vehicle lane far away from an opposite lane is detected;

the driving intention recognition module is used for inputting the state data into a driving intention recognition model trained in advance to obtain the driving intention of the target vehicle;

a target vehicle predicted trajectory generation module for generating a third predicted trajectory of the target vehicle based on the travel intention;

and the control module is used for controlling the main vehicle according to the third predicted track.

The automatic vehicle meeting control device provided by the embodiment of the invention can execute the automatic vehicle meeting control method provided by the first embodiment and the second embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.

Example four

Fig. 5 shows a schematic configuration of a vehicle control device 50 that can be used to implement the present invention. Vehicle control devices are intended to mean devices containing various forms of digital computers, such as devices containing laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers.

As shown in fig. 5, the vehicle control apparatus includes at least one processor 51, and a memory communicatively connected to the at least one processor 51, such as a Read Only Memory (ROM) 52, a Random Access Memory (RAM) 53, and the like, in which a computer program executable by the at least one processor is stored, and the processor 51 may perform various appropriate actions and processes according to the computer program stored in the Read Only Memory (ROM) 52 or the computer program loaded from a storage unit 58 into the Random Access Memory (RAM) 53. In the RAM 53, various programs and data necessary for the operation of the vehicle control apparatus 50 can also be stored. The processor 51, the ROM 52, and the RAM 53 are connected to each other via a bus 54. An input/output (I/O) interface 55 is also connected to bus 54.

A plurality of components in the vehicle control apparatus 50 are connected to the I/O interface 55, including: an input unit 56 such as a keyboard, a mouse, a sensor, and the like; an output unit 57 such as various types of displays, speakers, and the like; a storage unit 58 such as a magnetic disk, optical disk, or the like; and a communication unit 59 such as a network card, modem, wireless communication transceiver, etc. The communication unit 59 allows the vehicle control apparatus 50 to exchange information/data with other apparatuses via a computer network such as the internet and/or various telecommunication networks.

The processor 51 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of processors 51 include, but are not limited to, central Processing Units (CPUs), graphics Processing Units (GPUs), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processors, controllers, microcontrollers, and the like. The processor 51 performs the various methods and processes described above, such as an autonomous vehicle meeting control method.

In some embodiments, the autonomous vehicle crossing control method may be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 58. In some embodiments, part or all of the computer program may be loaded and/or installed onto the vehicle control apparatus 50 via the ROM 52 and/or the communication unit 59. When loaded into RAM 53 and executed by processor 51, may perform one or more of the steps of the above-described autonomous vehicle crossing control method, and/or detection method. Alternatively, in other embodiments, processor 41 may be configured to perform the autonomous vehicle crossing control method in any other suitable manner (e.g., by way of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), system on a chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

Computer programs for implementing the methods of the present invention can be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be performed. A computer program can execute entirely on a machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. A computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here may be implemented on a vehicle control device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user may provide input to the vehicle control apparatus. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.

The computing system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical host and VPS service are overcome.

It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present invention may be executed in parallel, sequentially, or in different orders, and are not limited herein as long as the desired results of the technical solution of the present invention can be achieved.

The above-described embodiments should not be construed as limiting the scope of the invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An automatic vehicle driving meeting control method is characterized by comprising the following steps:

acquiring a running track of the target vehicle;

2. The method of claim 1, further comprising, prior to generating first and second predicted trajectories for the target vehicle to travel from the intersection to the host and opposing lanes, respectively:

detecting whether an intersection exists at one side of a main vehicle lane far away from an opposite lane and in front of the main vehicle in the driving process of the main vehicle;

if yes, detecting whether a target vehicle with a vehicle head running towards the main vehicle lane exists at the intersection;

when there is a target vehicle whose head is traveling toward the host lane at the intersection, the step of generating a first predicted trajectory and a second predicted trajectory for the target vehicle to travel from the intersection to the host lane and the oncoming lane, respectively, is performed.

3. The method of claim 1, wherein generating a first predicted trajectory and a second predicted trajectory for a target vehicle to travel from an intersection to a host lane and to a oncoming lane, respectively, upon detecting the target vehicle at the intersection ahead of the host vehicle on a side of the host lane away from the oncoming lane comprises:

determining a first target position and a second target position, the first target position being a position after the target vehicle travels from the intersection to the host lane, the second target position being a position after the target vehicle travels from the intersection to the oncoming lane;

and respectively generating a first predicted track and a second predicted track of the target vehicle driving from the intersection to the main lane and the opposite lane by taking the current position of the target vehicle as an initial position, respectively taking the first target position and the second target position as end positions and taking the current state of the target vehicle as the input of a track planning algorithm.

4. The method according to any one of claims 1 to 3, wherein the travel locus is a locus acquired according to a preset period, the calculating of the first similarity of the travel locus to the first predicted locus and the calculating of the second similarity of the travel locus to the second predicted locus include:

acquiring the starting time and the ending time of the driving track;

respectively intercepting tracks from the starting time to the ending time from the first predicted track and the second predicted track to obtain a first sub-track and a second sub-track;

and calculating the similarity between the running track and the first sub-track to obtain a first similarity, and calculating the similarity between the running track and the second sub-track to obtain a second similarity.

5. The method of claim 4, wherein the calculating the similarity of the travel track and the first sub-track yields a first similarity, comprising:

acquiring an indication value corresponding to the state of a steering lamp of the target vehicle close to one side of the host vehicle from the starting time to the ending time;

respectively sampling the running track and the first sub-track according to a preset time interval to obtain a plurality of running track points on the running track and sub-track points on the first sub-track, wherein each running track point and each sub-track point comprise at least one item of state data;

calculating the similarity of the driving track point and the sub-track point at the same moment by the following formula:

calculating the sum of the similarity of the driving track points and the sub-track points at a plurality of sampling moments;

calculating the similarity of the driving track and the first sub-track by the following formula to obtain a first similarity:

in the formula, S is first similarity, a is weight of a sum value of the similarity of a plurality of track points calculated according to state data of k moments, C is an indication value of the state of the turn signal, C is smaller than 0 when the turn signal flickers, C is larger than 0 when the turn signal does not flick, and b is weight of the state of the turn signal.

6. The method of claim 1, wherein the determining a travel intention of the target vehicle based on the first similarity and the second similarity and controlling the host vehicle based on the travel intention comprises:

judging whether the first similarity and the second similarity are larger than a similarity threshold value or not;

determining that the travel intent of the target vehicle is to travel from the intersection to the host lane when the first similarity is greater than the similarity threshold;

controlling the host vehicle to run according to the first predicted trajectory and an original trajectory of the host vehicle, wherein the original trajectory is a running trajectory of the host vehicle when the target vehicle is not detected;

when the second similarity is larger than the similarity threshold value, determining that the driving intention of the target vehicle is to drive from the intersection to the opposite lane;

and controlling the main vehicle to stop running so as to avoid the target vehicle.

7. The method of claim 6, wherein the controlling the host vehicle to travel based on the first predicted trajectory and the original trajectory of the host vehicle comprises:

detecting whether there is a oncoming vehicle in the oncoming lane upon determining that there is an overlapping travel area between the host vehicle and the oncoming vehicle from the first predicted trajectory and the original trajectory;

if so, controlling the host vehicle to decelerate and driving the target vehicle to pass through the intersection along with the target vehicle after the target vehicle drives to the host vehicle lane;

if not, planning a target running track, wherein the target running track is a track returning to the main vehicle lane after passing through the opposite lane to avoid the target vehicle;

and controlling the main vehicle to run according to the target running track.

8. The method of claim 1 or 2, wherein when a target vehicle is detected at an intersection ahead of the host vehicle on a side of the host vehicle lane away from the oncoming lane, further comprising:

acquiring state data of the target vehicle within a preset time length;

inputting the state data into a pre-trained driving intention recognition model to obtain the driving intention of the target vehicle;

generating a third predicted trajectory of the target vehicle based on the travel intent;

and controlling the host vehicle according to the third predicted track.

9. An automatic vehicle meeting control device, characterized by comprising:

the similarity calculation module is used for calculating a first similarity between the driving track and the first predicted track and calculating a second similarity between the driving track and the second predicted track;

10. A vehicle control apparatus, characterized by comprising:

at least one processor; and

the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-8.

11. A computer readable storage medium, characterized in that it stores computer instructions for causing a processor, when executed, to implement the automated vehicle crossing control method of any one of claims 1-8.