Obstacle trajectory prediction method, device, equipment and storage medium

Abstract

The embodiment of the present invention discloses an obstacle trajectory prediction method, device, equipment and storage medium, the method comprising: obtaining first obstacle information corresponding to a first obstacle traveling on a structured road and second obstacle information corresponding to a second obstacle within a preset range of the first obstacle; determining a first driving guide line corresponding to the first obstacle according to the road map information of the structured road and the first obstacle information; predicting the target future trajectory of the first obstacle according to the first driving guide line, the first obstacle information and the second obstacle information. The technical solution of the embodiment of the present invention can improve the accuracy of obstacle trajectory prediction.

Description

Technical Field

[0001] The embodiments of the present invention relate to computer technology, and in particular to an obstacle trajectory prediction method, device, equipment and storage medium. Background Art

[0002] With the rapid development of computer technology, vehicles can achieve unmanned autonomous driving on structured roads. In unmanned autonomous driving scenarios, it is usually necessary to predict the movement trajectory of obstacles around the unmanned vehicle to avoid collisions between the unmanned vehicle and obstacles.

[0003] Currently, the future motion trajectory of the obstacle is usually obtained by linear interpolation based on the obstacle's speed and direction.

[0004] However, in the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art:

[0005] The obstacle information collected by the sensor may contain noise. For example, the collected driving direction of the obstacle is opposite to the actual driving direction, which makes the predicted future trajectory of the obstacle unreasonable, thereby reducing the accuracy of the obstacle trajectory prediction. Summary of the invention

[0006] Embodiments of the present invention provide an obstacle trajectory prediction method, apparatus, device and storage medium to improve the accuracy of obstacle trajectory prediction.

[0007] In a first aspect, an embodiment of the present invention provides an obstacle trajectory prediction method, comprising:

[0008] Acquire first obstacle information corresponding to a first obstacle traveling on a structured road and second obstacle information corresponding to a second obstacle within a preset range of the first obstacle;

[0009] Determining a first driving guide line corresponding to the first obstacle according to the road map information of the structured road and the first obstacle information;

[0010] A target future trajectory of the first obstacle is predicted according to the first driving guide line, the first obstacle information, and the second obstacle information.

[0011] In a second aspect, an embodiment of the present invention further provides an obstacle trajectory prediction device, comprising:

[0012] An obstacle information acquisition module, used to acquire first obstacle information corresponding to a first obstacle traveling on a structured road and second obstacle information corresponding to a second obstacle within a preset range of the first obstacle;

[0013] A first driving guide line determining module, configured to determine a first driving guide line corresponding to the first obstacle according to the road map information of the structured road and the first obstacle information;

[0014] The target future trajectory prediction module is used to predict the target future trajectory of the first obstacle according to the first driving guide line, the first obstacle information and the second obstacle information.

[0015] In a third aspect, an embodiment of the present invention further provides a device, the device comprising:

[0016] one or more processors;

[0017] A memory for storing one or more programs;

[0018] When the one or more programs are executed by the one or more processors, the one or more processors implement the obstacle trajectory prediction method as described in any embodiment of the present invention.

[0019] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the obstacle trajectory prediction method as described in any embodiment of the present invention.

[0020] The embodiments of the above invention have the following advantages or beneficial effects:

[0021] By using the first obstacle information corresponding to the first obstacle traveling on the structured road and the road map information of the structured road, the first driving guide line used to characterize the driving direction of the future trajectory of the first obstacle can be accurately determined, so that based on the first driving guide line and the second obstacle information of the second obstacle within the preset range of the first obstacle, the target future trajectory of the first obstacle can be accurately predicted, avoiding the situation where the trajectory prediction is unreasonable due to the noise in the collected obstacle information, and improving the accuracy of obstacle prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a flow chart of an obstacle trajectory prediction method provided by Embodiment 1 of the present invention;

[0023] Figure 2 is a flow chart of an obstacle trajectory prediction method provided by Embodiment 2 of the present invention;

[0024] Figure 3 is an example of a first driving guide line involved in the second embodiment of the present invention;

[0025] Figure 4 is a flow chart of an obstacle trajectory prediction method provided by Embodiment 3 of the present invention;

[0026] Figure 5 is an example of a reference point to be driven involved in the third embodiment of the present invention;

[0027] Figure 6 is a schematic diagram of the structure of an obstacle trajectory prediction device provided by Embodiment 4 of the present invention;

[0028] Figure 7 It is a structural schematic diagram of a device provided in Embodiment 5 of the present invention. DETAILED DESCRIPTION

[0029] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0030] Embodiment 1

[0031] Figure 1 This is a flowchart of obstacle trajectory prediction provided in Embodiment 1 of the present invention. This embodiment is applicable to the case of predicting the future trajectory of obstacles around an autonomous driving vehicle on a structured road. The method can be executed by an obstacle trajectory prediction device, which can be implemented by software and / or hardware and integrated into a device with an autonomous driving function, such as any type of vehicle. Figure 1 As shown, the method specifically comprises the following steps:

[0032] S110: Acquire first obstacle information corresponding to a first obstacle traveling on a structured road and second obstacle information corresponding to a second obstacle within a preset range of the first obstacle.

[0033] Among them, the structured road may refer to a road with traffic driving rules. Exemplarily, the structured road may include one or more lanes, each lane having traffic driving rules such as a prescribed vehicle driving direction. For example, an urban road may be used as a structured road. The first obstacle may refer to any vehicle around the autonomous driving vehicle traveling on the structured road. The first obstacle may refer to an obstacle in a structured road that is not in an intersection. The first obstacle and the autonomous driving vehicle may travel on the structured road at the same time. The preset range of the first obstacle may be a pre-set road area range that will affect the driving trajectory of the first obstacle. For example, the first preset range may be set to: a circular area range with the current position of the first obstacle as the center and a preset distance as the radius. The second obstacle may refer to each obstacle traveling on the structured road within the preset range of the first obstacle. The number of second obstacles may be one or more. For example, the second obstacle may refer to each vehicle around the first obstacle. The autonomous driving vehicle in this embodiment may serve as the second obstacle around the first obstacle. The first obstacle information may include but is not limited to: the current position information of the first obstacle, the driving direction of the first obstacle, the driving speed of the first obstacle, and the volume size of the first obstacle. The second obstacle information may include, but is not limited to: current position information of the second obstacle, the travel direction of the second obstacle, the travel speed of the second obstacle, and the volume of the second obstacle.

[0034] Specifically, the sensor can be used to collect information about each obstacle in the structured road. According to the current position of the autonomous driving vehicle and the collected current position information of each obstacle, any obstacle around the autonomous driving vehicle is taken as a first obstacle, and each second obstacle located within a preset range of the first obstacle is determined from the collected obstacles, so that the first obstacle information and the second obstacle information can be obtained.

[0035] S120: Determine a first driving guide line corresponding to the first obstacle according to the road map information of the structured road and the first obstacle information.

[0036] Among them, the road map information may refer to the information of lane driving rules in the structured road. For example, the road map information may include but is not limited to: lane information corresponding to each lane in the structured road and the lane line type between two adjacent lanes. Among them, the lane line type may include but is not limited to: single solid line, single dashed line and double solid line. The lane line type can be used to characterize whether the vehicle can cross between two adjacent lanes. For example, when the lane line type is a single solid line or a double solid line, the two adjacent lanes are non-crossable lanes. When the lane line type is a single dashed line, the two adjacent lanes are crossable lanes. Lane information may include but is not limited to: driving reference direction, lane centerline coordinate information, lane length and lane width. The driving reference direction may refer to the vehicle driving direction specified in the traffic driving rules. The first driving guide line corresponding to the first obstacle may be a guide line used to characterize the future trajectory driving direction and driving lane of the first obstacle.

[0037] Specifically, the collected first obstacle information can be corrected based on the road map information of the structured road, so that the first driving guide line that conforms to the actual driving information of the first obstacle can be accurately determined, avoiding the situation where unreasonable trajectory prediction is caused by inaccurate obstacle information collected by the sensor, thereby improving the accuracy of trajectory prediction.

[0038] Exemplarily, S120 may include: determining a current lane in which the first obstacle is located according to road map information of the structured road and current position information of the first obstacle; and determining a first driving guide line corresponding to the first obstacle according to the current lane.

[0039] Specifically, the current lane where the first obstacle is located at the current moment can be determined based on the lane information corresponding to each lane in the road map information of the structured road and the current position information of the first obstacle. The first driving guide line corresponding to the first obstacle can be determined based on the first driving reference direction of the current lane and the lane centerline coordinate information. For example, the lane centerline of the current lane can be used as the first driving guide line, and the first driving reference direction of the current lane can be used as the guiding direction of the first driving guide line. If the currently collected driving direction of the first obstacle is different from the first driving reference direction of the current lane, it indicates that the currently collected driving direction of the first obstacle is incorrect, so that a more accurate first driving guide line can be determined based on the current lane, avoiding the situation in the prior art where the first driving guide line is determined directly based on the collected driving direction of the first obstacle, resulting in an unreasonable predicted future trajectory, thereby improving the trajectory prediction accuracy.

[0040] S130: predicting a target future trajectory of the first obstacle according to the first driving guide line, the first obstacle information, and the second obstacle information.

[0041] Specifically, based on the first driving guide line and the second obstacle information around the first obstacle, the target future trajectory of the first obstacle can be accurately predicted. Each first obstacle around the autonomous driving vehicle can predict the corresponding target future trajectory using the above steps S110-S130, so that the autonomous driving vehicle can make trajectory decisions and plans based on the predicted target future trajectory of each first obstacle, obtain a reasonable vehicle driving path, and avoid collisions between the autonomous driving vehicle and the first obstacle, as well as unnecessary jamming and detours of the autonomous driving vehicle.

[0042] Exemplarily, S130 may include: determining various candidate future trajectories of the first obstacle in the structured road according to the first driving guide line; and determining a target future trajectory of the first obstacle from the candidate future trajectories according to the first obstacle information and the second obstacle information.

[0043] The candidate future trajectory may refer to a motion trajectory of the first obstacle that can be driven on the structured road. There may be multiple candidate future trajectories. The target future trajectory may be the final future trajectory selected from the candidate future trajectories.

[0044] Specifically, based on the lane and the guidance direction of the first driving guide line, various candidate future trajectories of the first obstacle can be determined in the structured road, and according to the first obstacle information and the second obstacle information, various candidate future trajectories can be screened, so as to obtain a more reasonable target future trajectory, further improving the trajectory prediction accuracy.

[0045] The technical solution of this embodiment can accurately determine a first driving guide line used to characterize the driving direction of the future trajectory of the first obstacle based on the first obstacle information corresponding to the first obstacle traveling on the structured road and the road map information of the structured road, so that based on the first driving guide line and the second obstacle information of the second obstacle within the preset range of the first obstacle, the target future trajectory of the first obstacle can be accurately predicted, thereby avoiding the situation where unreasonable trajectory prediction is caused by noise in the collected obstacle information, and improving the accuracy of obstacle prediction.

[0046] Embodiment 2

[0047] Figure 2 This is a flow chart of an obstacle trajectory prediction method provided by Embodiment 2 of the present invention. Based on the above embodiment, this embodiment describes in detail the method for determining the first driving guide line. The explanations of the terms that are the same or corresponding to the above embodiment are not repeated here.

[0048] See also Figure 2 The obstacle trajectory prediction method provided in this embodiment specifically includes the following steps:

[0049] S210: Acquire first obstacle information corresponding to a first obstacle traveling on a structured road and second obstacle information corresponding to a second obstacle within a preset range of the first obstacle.

[0050] S220: Determine a current lane in which the first obstacle is located according to the road map information of the structured road and the first obstacle position information of the first obstacle.

[0051] The current lane may refer to the lane where the first obstacle in the structured road is currently located.

[0052] Specifically, position matching may be performed according to lane information corresponding to each lane in the road map information of the structured road and the current position information of the first obstacle to determine the current lane where the first obstacle is located at the current moment.

[0053] S230, detecting whether the first driving reference direction of the current lane is consistent with the first obstacle driving direction of the first obstacle, if so, proceeding to step S240; if not, proceeding to step S250.

[0054] The first driving reference direction may refer to the driving direction of the current lane specified by the traffic driving rules. The first obstacle driving direction may be the driving direction of the first obstacle collected by the sensor.

[0055] Specifically, it may be detected whether the first driving reference direction of the current lane is consistent with the first obstacle driving direction of the first obstacle, so as to determine whether the collected first obstacle driving direction complies with the traffic driving rules.

[0056] S240. Determine a first driving guide line corresponding to a first obstacle according to a first lane center line of a current lane and a first driving reference direction, and proceed to step S260.

[0057] Specifically, if the first driving reference direction of the current lane is consistent with the first obstacle driving direction of the first obstacle, it means that the driving direction of the first obstacle complies with the traffic driving rules, the collected driving direction of the first obstacle is accurate, and the current lane can be used as the real driving lane of the first obstacle. At this time, the first lane centerline of the current lane can be used as the first driving guide line, and the first driving reference direction of the current lane or the first obstacle driving direction can be used as the guiding direction of the first driving guide line, so as to determine the first driving guide line corresponding to the first obstacle. For example, Figure 3 An example of a first driving guide line is given. Figure 33(a) in the figure, the structured road includes two lanes lane1 and lane2, the current lane where the first obstacle obs is located is lane2, and the first driving reference direction R1 of the current lane lane2 is consistent with the first obstacle driving direction R2 of the first obstacle, that is, both point upward. At this time, the lane centerline of the current lane lane2 can be used as the first driving guidance line, and the first driving reference direction of the current lane lane2 or the first obstacle driving direction can be used as the guidance direction of the first driving guidance line.

[0058] S250, detecting whether there is a crossable lane adjacent to the current lane, and determining a first driving guide line corresponding to the first obstacle according to the detection result, and proceeding to step S260.

[0059] The crossable lane may refer to an adjacent lane when the behavior of the first obstacle moving from the current lane to the adjacent lane complies with traffic driving rules.

[0060] Specifically, if the first driving reference direction of the current lane is inconsistent with the first obstacle driving direction of the first obstacle, it is possible to further detect whether there is a crossable lane adjacent to the current lane in the structured road, so as to further determine whether the driving direction of the first obstacle is really not in compliance with traffic driving rules, that is, whether the collected driving direction of the first obstacle is really wrong, so that based on the detection result, the first driving guide line of the first obstacle can be more accurately determined, thereby further improving the accuracy of the predicted trajectory.

[0061] Exemplarily, step S250 may specifically include the following steps S251-S253:

[0062] S251. Check whether there is a crossable lane adjacent to the current lane. If so, proceed to step S252; if not, proceed to step S253.

[0063] S252: Determine a first driving guide line corresponding to the first obstacle according to the second driving reference direction that can cross the lane and the driving direction of the first obstacle, and proceed to step S260.

[0064] Specifically, if a crossable lane adjacent to the current lane is detected in the structured road, the first driving guide line corresponding to the first obstacle can be accurately determined according to the second driving reference direction of the crossable lane and the driving direction of the first obstacle.

[0065] Exemplarily, S252 may include: detecting whether the second driving reference direction of the crossable lane is consistent with the driving direction of the first obstacle; if consistent, determining the first driving guide line corresponding to the first obstacle based on the second lane center line of the crossable lane and the second driving reference direction; if inconsistent, determining the first driving guide line corresponding to the first obstacle based on the first lane center line of the current lane and the first driving reference direction.

[0066] Specifically, if it is detected that the second driving reference direction of the crossable lane is consistent with the driving direction of the first obstacle, it means that the first obstacle is in an overtaking state and temporarily occupies the current lane, which complies with traffic driving rules. The crossable lane can be used as the real driving lane of the first obstacle. At this time, the center line of the second lane of the crossable lane can be used as the first driving guide line, and the driving direction of the first obstacle or the second driving reference direction of the crossable lane can be used as the guiding direction of the first driving guide line. For example, Figure 3 3(c) in the figure, the structured road includes two lanes lane1 and lane2, the current lane where the first obstacle obs is located is lane1, and there is a crossable lane lane2, the second driving reference direction R3 of the crossable lane lane2 is consistent with the first obstacle driving direction R2, that is, both point upwards, at this time, the lane centerline of the crossable lane lane2 can be used as the first driving guide line, and the second driving reference direction of the crossable lane lane2 or the first obstacle driving direction can be used as the guiding direction of the first driving guide line.

[0067] If it is detected that the second driving reference direction of the crossable lane is inconsistent with the driving direction of the first obstacle, that is, the driving direction of the first obstacle is different from the first driving reference direction of the current lane and the second driving reference direction of the crossable lane, it means that the driving direction of the first obstacle does not comply with the traffic driving rules, and the collected driving direction of the first obstacle is wrong. At this time, the center line of the first lane of the current lane can be used as the first driving guide line, and the first driving reference direction of the current lane can be used as the guiding direction of the first driving guide line, so as to more accurately determine the first driving guide line corresponding to the first obstacle. For example, Figure 3 3(d) in the figure, the structured road includes two lanes lane1 and lane2, the current lane where the first obstacle obs is located is lane2, and there is a lane that can be crossed over lane1, and the second driving reference direction R3 that can be crossed over lane1 is opposite to the driving direction R2 of the first obstacle. At this time, the lane centerline of the current lane lane2 can be used as the first driving guidance line, and the first driving reference direction of the current lane lane2 can be used as the guidance direction of the first driving guidance line.

[0068] Exemplarily, S252 may also include: if there are at least two second driving reference directions of crossable lanes that are consistent with the driving direction of the first obstacle, determining the target crossable lane from each crossable lane according to the distance between each crossable lane and the first obstacle; determining the first driving guide line corresponding to the first obstacle according to the second lane center line of the target crossable lane and the second driving reference direction of the target crossable lane.

[0069] Specifically, if it is detected that there are multiple crossable lanes in the structured road, it is possible to further detect whether the second driving reference direction of each crossable lane is the same as the driving direction of the first obstacle. If it is detected that there are at least two crossable lanes with the same driving direction as the first obstacle, the distance between the centerline position of each crossable lane and the current position of the first obstacle can be determined. The smaller the distance, the greater the possibility of crossing the first obstacle. Thus, the crossable lane with the smallest distance can be determined as the target crossable lane. The centerline of the second lane of the target crossable lane is used as the first driving guide line, and the second driving reference direction of the target crossable lane is used as the guiding direction of the first driving guide line, so as to determine a more accurate first driving guide line, further improving the accuracy of trajectory prediction.

[0070] S253. Determine a first driving guide line corresponding to the first obstacle according to the first lane center line of the current lane and the first driving reference direction, and proceed to step S260.

[0071] Specifically, if it is detected that there is no crossable lane adjacent to the current lane in the structured road, it means that the first obstacle has not crossed the lane to overtake, the current lane can be used as the real driving lane of the first obstacle, and the driving direction of the first obstacle does not comply with the traffic driving rules. The collected driving direction of the first obstacle is wrong. At this time, the center line of the first lane of the current lane can be used as the first driving guide line, and the first driving reference direction of the current lane can be used as the guiding direction of the first driving guide line, so as to determine the accurate first driving guide line. For example, Figure 3 3(b) in the figure, the structured road includes two lanes lane1 and lane2, the current lane where the first obstacle obs is located is lane2, the first driving reference direction R1 of the current lane lane2 is inconsistent with the driving direction R2 of the first obstacle, and there is no crossable lane in the current lane lane2. At this time, the lane centerline of the current lane lane2 can be used as the first driving guide line, and the first driving reference direction of the current lane lane2 can be used as the guiding direction of the first driving guide line.

[0072] S260: predicting a target future trajectory of the first obstacle according to the first driving guide line, the first obstacle information, and the second obstacle information.

[0073] The technical solution of this embodiment is to determine the first driving guide line corresponding to the first obstacle according to the first lane center line and the first driving reference direction of the current lane when it is detected that the first driving reference direction of the current lane is consistent with the first obstacle driving direction of the first obstacle; when it is detected that the first driving reference direction of the current lane is inconsistent with the first obstacle driving direction of the first obstacle, further detect whether there is a crossable lane adjacent to the current lane, and determine the first driving guide line corresponding to the first obstacle according to the detection result, so that a more accurate first driving guide line can be obtained, and the accuracy of obstacle trajectory prediction is further improved.

[0074] Embodiment 3

[0075] Figure 4 This is a flow chart of an obstacle trajectory prediction method provided in the third embodiment of the present invention. Based on the above embodiments, this embodiment optimizes "determining each candidate future trajectory of the first obstacle in the structured road according to the first driving guide line". On this basis, it further optimizes "determining the target future trajectory of the first obstacle from the candidate future trajectory according to the first obstacle information and the second obstacle information". The explanations of the terms that are the same or corresponding to the above embodiments are not repeated here.

[0076] See also Figure 4 The obstacle trajectory prediction method provided in this embodiment specifically includes the following steps:

[0077] S310: Acquire first obstacle information corresponding to a first obstacle traveling on a structured road and second obstacle information corresponding to a second obstacle within a preset range of the first obstacle.

[0078] S320: Determine a first driving guide line corresponding to the first obstacle according to the road map information of the structured road and the first obstacle information.

[0079] S330: Determine a target to-be-traveled area of ​​the first obstacle in the structured road according to the first driving guide line.

[0080] The target area to be driven may refer to the road area where the first obstacle is to be driven in the structured road. Specifically, the target area to be driven of the first obstacle may be accurately determined based on the lane and the guiding direction of the first driving guide line. Exemplarily, S330 may include: determining the first area to be driven in front of the first obstacle in the current lane where the first obstacle is located based on the guiding direction of the first driving guide line and the preset area length; determining the target area to be driven of the first obstacle in the structured road according to the first area to be driven.

[0081] The preset area length may be a road length that is set in advance based on business needs and that the first obstacle will travel within a preset future period of time. Specifically, starting from the current position of the first obstacle, along the guidance direction of the first driving guide line, a road area of ​​the preset area length in front of the first obstacle in the current lane may be determined as the first waiting area for driving of the first obstacle. The first waiting area for driving may be directly determined as the target waiting area for driving of the first obstacle.

[0082] Exemplarily, determining the target area to be driven in the structured road of the first obstacle based on the first area to be driven can include: if it is detected that there is a crossable lane adjacent to the current lane, then based on the guidance direction of the first driving guide line and the preset area length, determining the second area to be driven in front of the first obstacle in the crossable lane; determining the target area to be driven in the structured road of the first obstacle based on the first area to be driven and the second area to be driven.

[0083] Specifically, by detecting whether there is a crossable lane adjacent to the current lane in the structured road, other areas to be driven where the first obstacle may drive can be determined based on the detection result, so as to improve the accuracy of determining the target area to be driven and further improve the accuracy of trajectory prediction. If it is detected that there is a crossable lane adjacent to the current lane, it means that the first obstacle will not only drive in the current lane, but may also cross to the crossable lane. At this time, the road area with a preset area length in front of the first obstacle in the crossable lane can be determined as the second area to be driven of the first obstacle along the guidance direction of the first driving guide line with the current position of the first obstacle as the starting point. The total area surrounded by the first area to be driven and the second area to be driven can be determined as the target area to be driven of the first obstacle in the structured road. If it is detected that there is no crossable lane adjacent to the current lane, it means that the first obstacle will only drive in the current lane. At this time, the first area to be driven can be directly determined as the target area to be driven of the first obstacle, thereby further improving the accuracy of the target area to be driven.

[0084] It should be noted that, when there is a crossable lane adjacent to the current lane, if the regional lane line corresponding to the second area to be driven includes both a dashed line and a solid line, it means that the first obstacle can only be crossed from the dashed line, but not from the solid line. At this time, the second area to be driven can be divided into regions based on the solid line and the dashed line to obtain the third area to be driven corresponding to the single dashed line. Among them, the regional lane line may refer to a lane line between the current lane and the crossable lane of a preset regional length in the second area to be driven. In view of the above situation, the total area surrounded by the first area to be driven and the third area to be driven can be determined as the target area to be driven of the first obstacle in the structured road, further improving the accuracy of the target area to be driven.

[0085] S340, dividing the target area to be driven into regions, and determining reference points to be driven on the boundary lines of each layer.

[0086] The reference point to be driven may refer to a possible position of the first obstacle when the first obstacle drives to the corresponding dividing line. Each layer of dividing line may include multiple reference points to be driven.

[0087] Specifically, the target area to be driven can be layered into equal distance areas based on the first preset interval distance, so that the distance between each two adjacent layers of dividing lines is equal, thereby further simplifying the trajectory prediction operation. Among them, the first preset interval distance can be determined based on the driving speed of the first obstacle and the preset unit time. For example, the product of the driving speed of the first obstacle and the preset unit time can be determined as the first preset interval distance. After determining each layer of dividing lines, each position point on each layer of dividing lines can be randomly selected as a reference point to be driven; or each position point on each layer of dividing lines separated by a second preset interval distance can be determined as a reference point to be driven, so as to ensure that the distance between each two adjacent reference points to be driven is equal, thereby further simplifying the trajectory prediction operation.

[0088] For example, Figure 5 An example of a reference point to be traveled is given. Figure 5 The positions of the four reference points to be driven in Figure 3 The determination situations of the four first guide lines in correspondence one by one. For example, as shown in 5(a) in 5, when the first area to be driven in lane 2 and the second area to be driven in lane 1 form a target area to be driven, the determined reference points to be driven are shown as the dots in 5(a). As shown in 5(b) in 5, when the first area to be driven in lane 2 is used as the target area to be driven, the determined reference points to be driven are shown as the dots in 5(b). As shown in 5(c) in 5, when the first area to be driven in lane 1 and the third area to be driven in lane 2 form a target area to be driven, the determined reference points to be driven are shown as the dots in 5(c). As shown in 5(d) in 5, when the first area to be driven in lane 2 and the third area to be driven in lane 1 form a target area to be driven, the determined reference points to be driven are shown as the dots in 5(d).

[0089] S350: Determine each candidate future trajectory corresponding to the first obstacle according to the reference point to be driven.

[0090] Specifically, a reference point to be traveled can be selected from all the reference points to be traveled on each layer of the dividing line, and a corresponding future trajectory to be selected can be determined based on each selected reference point to be traveled. For example, each selected reference point to be traveled can be connected to obtain a corresponding future trajectory to be selected, so that all different future trajectories to be selected in the target area to be traveled can be obtained.

[0091] S360: Determine a trajectory cost value corresponding to each future trajectory to be selected according to the first obstacle information and the second obstacle information.

[0092] The trajectory cost value can be used to characterize the unreasonableness of each future trajectory prediction. For example, the trajectory cost value can be used to specifically characterize the probability and discomfort of a first obstacle colliding with a second obstacle when traveling along the future trajectory. Specifically, the trajectory cost value of the first obstacle when traveling along each future trajectory can be determined based on the first obstacle information and the second obstacle information.

[0093] Exemplarily, S360 may include: determining a current reference point cost value corresponding to a current reference point to be traveled in the selected future trajectory according to the first obstacle information, the second obstacle information and the previous reference point cost value corresponding to the previous reference point to be traveled in the selected future trajectory; determining a trajectory cost value corresponding to the selected future trajectory according to the reference point cost value corresponding to each reference point to be traveled in the selected future trajectory.

[0094] Among them, the reference point cost value corresponding to the reference point to be traveled can be used to characterize the unreasonable degree of the first obstacle when traveling to the reference point to be traveled. Specifically, for each reference point to be traveled in the future trajectory to be selected, the current reference point cost value corresponding to the current reference point to be traveled in the future trajectory to be selected can be determined based on the first obstacle information, the second obstacle information and the previous reference point cost value corresponding to the previous reference point to be traveled. It should be noted that if the current reference point to be traveled is the first reference point to be traveled in the future trajectory to be selected, the previous reference point cost value corresponding to the corresponding previous reference point to be traveled can be set to 0. For each future trajectory to be selected, the reference point cost values ​​corresponding to each reference point to be traveled in the future trajectory to be selected can be added, and the addition result can be determined as the trajectory cost value corresponding to the future trajectory to be selected. Similarly, the trajectory cost value corresponding to each future trajectory to be selected can be obtained.

[0095] S370: Determine a target future trajectory of the first obstacle according to the trajectory cost value corresponding to each candidate future trajectory.

[0096] Specifically, the trajectory cost values ​​corresponding to each candidate future trajectory are compared. The higher the trajectory cost value, the more unreasonable the candidate future trajectory is. Therefore, the candidate future trajectory with the smallest trajectory cost value can be determined as the target future trajectory, so that the accurate target future trajectory can be predicted, further improving the accuracy of trajectory prediction.

[0097] The technical solution of this embodiment can more accurately determine the target area to be driven by the first obstacle in the structured road according to the first driving guide line, and divide the target area to be driven into regions, and obtain more accurate future trajectories to be selected according to the reference points to be driven on the boundary line of each layer. By more accurately determining the target future trajectory of the first obstacle according to the trajectory cost value corresponding to each future trajectory to be selected, the accuracy of obstacle trajectory prediction is further improved.

[0098] On the basis of the above technical solution, determining the current reference point cost value corresponding to the current reference point to be driven in the selected future trajectory according to the first obstacle information, the second obstacle information and the previous reference point cost value corresponding to the previous reference point to be driven in the selected future trajectory may include:

[0099] According to the first obstacle information and the second obstacle information, a first generation value of the impact of the second obstacle on the first obstacle when the first obstacle travels from the previous reference point to be traveled in the selected future trajectory to the current reference point to be traveled is determined; according to the first driving guide line and the first driving path between the previous reference point to be traveled and the current reference point to be traveled, a second generation value of the offset of the first driving path is determined; a third generation value of the non-smoothness of the first driving path is determined; according to the previous reference point cost value, the first generation value, the second generation value and the third generation value corresponding to the previous reference point to be traveled, a current reference point cost value corresponding to the current reference point to be traveled in the selected future trajectory is determined.

[0100] Among them, the first generation value can be used to characterize the degree of influence of the surrounding second obstacles on the driving of the first obstacle when the first obstacle travels from the previous reference point to be traveled to the current reference point to be traveled. The first driving path may refer to the path obtained by connecting the previous reference point to be traveled and the current reference point to be traveled. Both the second generation value and the third generation value can be used to characterize the degree of discomfort of the first obstacle when traveling from the previous reference point to be traveled to the current reference point to be traveled. The second generation value can be used to characterize the degree of deviation of the first driving path. The third generation value can be used to characterize the degree of excessive curvature of the first driving path, that is, the degree of unevenness.

[0101] Specifically, when the first obstacle travels from the last reference point to be traveled in the selected future trajectory to the current reference point to be traveled, the first generation value of the impact of the second obstacle on the first obstacle can be obtained by determining the distance between the first obstacle and the second obstacle based on the first obstacle information and the second obstacle information. The second generation value of the deviation of the first driving path can be obtained by determining the distance or angle difference between the first driving guide line and the first driving path. The third generation value of the non-smoothness of the first driving path can be obtained by determining the slope, curvature or curvature change rate of the first driving path. The first cost weight value corresponding to the first generation value, the second cost weight value corresponding to the second generation value, and the third cost weight value corresponding to the third generation value can be preset, and the sum of the first cost weight value, the second cost weight value, and the third cost weight value is 1, so that the first generation value, the second generation value, and the third generation value can be weighted and summed based on the first cost weight value, the second cost weight value, and the third cost weight value, and the weighted summation result is added to the previous reference point cost value corresponding to the previous reference point to be driven, and the obtained addition result can be determined as the current reference point cost value corresponding to the current reference point to be driven. Similarly, the above method can be used to obtain the reference point cost value corresponding to each reference point to be driven in the selected future trajectory.

[0102] Exemplarily, determining, based on the first obstacle information and the second obstacle information, a first generation value of the impact of the second obstacle on the first obstacle when the first obstacle travels from the previous reference point to be traveled in the selected future trajectory to the current reference point to be traveled, may include:

[0103] According to the first obstacle information and the second obstacle information, a collision cost value of a collision between the first obstacle and the first obstacle when the first obstacle travels from the last reference point to be traveled in the selected future trajectory to the current reference point to be traveled, and when the second obstacle travels from the first reference point to be traveled to the second reference point to be traveled, is determined, wherein the first reference point to be traveled is a reference point to be traveled corresponding to a second obstacle at the same layer as the last reference point to be traveled, and the second reference point to be traveled is a reference point to be traveled corresponding to a second obstacle at the same layer as the current reference point to be traveled; according to the second driving guide line corresponding to the second obstacle and the second driving path between the first reference point to be traveled and the second reference point to be traveled, an offset cost value of an offset of the second driving path is determined; a path cost value of a non-smooth second driving path is determined; and according to the collision cost value, the offset cost value and the path cost value, a first cost value of an impact of the second obstacle on the first obstacle is determined.

[0104] Among them, the various reference points to be traveled of the second obstacle on each layer of the boundary line can be determined based on a method similar to that of determining the reference points to be traveled corresponding to the first obstacle. The first reference point to be traveled can refer to a reference point to be traveled that is arbitrarily selected on the boundary line of the second obstacle that is on the same layer as the previous reference point to be traveled. The second reference point to be traveled can refer to a reference point to be traveled that is arbitrarily selected on the boundary line of the second obstacle that is on the same layer as the current reference point to be traveled. The second driving guide line corresponding to the second obstacle can be determined based on the method of determining the first driving guide line corresponding to the first obstacle. The second driving path can refer to the path obtained by connecting the first reference point to be traveled and the second reference point to be traveled.

[0105] Among them, the collision cost value can be used to characterize the degree of influence of the second obstacle on the travel of the first obstacle when the first obstacle travels from the previous reference point to be traveled in the selected future trajectory to the current reference point to be traveled, and when the second obstacle travels from the first reference point to be traveled to the second reference point to be traveled. The offset cost value can be used to characterize the degree of offset of the second travel path corresponding to the second obstacle. The path cost value can be used to characterize the degree of excessive curvature of the second travel path, that is, the degree of unevenness. Both the offset cost value and the path cost value can be used to characterize the degree of discomfort of the second obstacle when it travels from the first reference point to be traveled to the second reference point to be traveled.

[0106] Specifically, for each second obstacle, when the first obstacle travels from the last reference point to be traveled in the selected future trajectory to the current reference point to be traveled, and the second obstacle travels from the first reference point to be traveled to the second reference point to be traveled, the collision cost value of the second obstacle colliding with the first obstacle can be obtained by determining the distance between the first travel path and the second travel path. The offset cost value of the second travel path offset can be obtained by determining the distance or angle difference between the second travel guide line corresponding to the second obstacle and the second travel path. The non-smooth path cost value of the second travel path can be obtained by determining the slope, curvature or curvature change rate of the second travel path. The collision cost weight value corresponding to the collision cost value, the offset cost weight value corresponding to the offset cost value and the path cost weight value corresponding to the path cost value can be pre-set, and the sum of the collision cost weight value, the offset cost weight value and the path cost weight value is 1, so that the first cost value of the impact of the second obstacle on the first obstacle can be determined based on the collision cost weight value, the offset cost weight value, the path cost weight value, the collision cost value, the offset cost value and the path cost value. For example, the first generation value of the impact of the second obstacle on the first obstacle can be determined by the following formula:

[0107]

[0108] in, It is the first generation value of the impact of all the surrounding second obstacles on the first obstacle when the first obstacle k travels from the previous reference point m to be traveled in the selected future trajectory to the current reference point n to be traveled. It is the collision cost value of the first obstacle k traveling from the previous reference point m to the current reference point n in the selected future trajectory, and the second obstacle p traveling from the first reference point u to the second reference point v, when the second obstacle p collides with the first obstacle k. It is the offset cost value of the second driving path when the second obstacle p travels from the first reference point u to be traveled to the second reference point v to be traveled. It is the path cost value of the non-smooth second driving path when the second obstacle p travels from the first reference point u to be traveled to the second reference point v to be traveled. and They are respectively the collision cost weight value, the offset cost weight value and the path cost weight value. Specifically, the collision cost value, the offset cost value and the path cost value can be weighted and summed, and the weighted sum result can be used as the impact cost value corresponding to the second obstacle p. By reselecting the first reference point u to be driven and the second reference point v to be driven, the impact cost value corresponding to the second obstacle p is minimized to obtain the minimum impact cost value. Similarly, based on the above method, the minimum impact cost value corresponding to each second obstacle can be obtained, and the various minimum impact cost values ​​can be added together. The obtained addition result can be determined as the first generation value of the impact of the second obstacle on the first obstacle, so that the first generation value corresponding to the first obstacle can be determined more accurately, further improving the accuracy of obstacle trajectory prediction.

[0109] The following is an embodiment of an obstacle trajectory prediction device provided by an embodiment of the present invention. The device and the obstacle trajectory prediction methods of the above embodiments belong to the same inventive concept. For details not described in detail in the embodiment of the obstacle trajectory prediction device, reference can be made to the embodiment of the above obstacle trajectory prediction method.

[0110] Embodiment 4

[0111] Figure 6 This is a schematic diagram of the structure of an obstacle trajectory prediction device provided in the fourth embodiment of the present invention. This embodiment can be applied to the situation of predicting the future trajectory of obstacles around an autonomous driving vehicle on a structured road. The device specifically includes: an obstacle information acquisition module 410, a first driving guide line determination module 420 and a target future trajectory prediction module 430.

[0112] Among them, the obstacle information acquisition module 410 is used to obtain the first obstacle information corresponding to the first obstacle traveling on the structured road and the second obstacle information corresponding to the second obstacle within the preset range of the first obstacle; the first driving guide line determination module 420 is used to determine the first driving guide line corresponding to the first obstacle based on the road map information of the structured road and the first obstacle information; the target future trajectory prediction module 430 is used to predict the target future trajectory of the first obstacle based on the first driving guide line, the first obstacle information and the second obstacle information.

[0113] Optionally, the first driving guide line determining module 420 includes:

[0114] a current lane determination submodule, configured to determine a current lane in which the first obstacle is located according to the road map information of the structured road and the first obstacle current position information of the first obstacle;

[0115] The first driving guide line determination submodule is used to determine the first driving guide line corresponding to the first obstacle according to the current lane.

[0116] Optionally, the first driving guide line determination submodule includes:

[0117] a driving direction detection unit, used to detect whether a first driving reference direction of a current lane is consistent with a first obstacle driving direction of a first obstacle;

[0118] A first determining unit, configured to determine, if the first obstacle is consistent, a first driving guide line corresponding to the first obstacle according to the first lane center line of the current lane and the first driving reference direction;

[0119] The second determining unit is used to detect whether there is a crossable lane adjacent to the current lane if there is no inconsistency, and determine the first driving guide line corresponding to the first obstacle according to the detection result.

[0120] Optionally, the second determining unit includes:

[0121] A first determining subunit is configured to determine, if there is a crossable lane adjacent to the current lane, a first driving guide line corresponding to the first obstacle according to a second driving reference direction of the crossable lane and a driving direction of the first obstacle;

[0122] The second determining subunit is used to determine a first driving guide line corresponding to the first obstacle according to a first lane center line and a first driving reference direction of the current lane if there is no crossable lane adjacent to the current lane.

[0123] Optionally, the first determination subunit is specifically used to: detect whether the second driving reference direction of the crossable lane is consistent with the driving direction of the first obstacle; if consistent, determine the first driving guide line corresponding to the first obstacle according to the second lane center line of the crossable lane and the second driving reference direction; if inconsistent, determine the first driving guide line corresponding to the first obstacle according to the first lane center line of the current lane and the first driving reference direction.

[0124] Optionally, the target future trajectory prediction module 430 includes:

[0125] A candidate future trajectory determination submodule, used to determine each candidate future trajectory of the first obstacle in the structured road according to the first driving guide line;

[0126] The target future trajectory determination submodule is used to determine the target future trajectory of the first obstacle from the candidate future trajectories according to the first obstacle information and the second obstacle information.

[0127] Optionally, the submodule for determining the future trajectory to be selected includes:

[0128] a target waiting driving area determining unit, configured to determine a target waiting driving area of ​​a first obstacle in a structured road according to a first driving guide line;

[0129] A to-be-traveled reference point determination unit is used to stratify the target to-be-traveled area and determine the to-be-traveled reference point on the boundary line of each layer;

[0130] The unit for determining a future trajectory to be selected is used to determine each future trajectory to be selected corresponding to the first obstacle according to the reference point to be driven.

[0131] Optionally, the target to-be-traveled area determination unit includes:

[0132] A first waiting-to-travel area determination subunit is used to determine a first waiting-to-travel area located in front of the first obstacle in the current lane where the first obstacle is located based on the guidance direction of the first driving guidance line and the preset area length;

[0133] The target to-be-traveled area determination subunit is used to determine a target to-be-traveled area of ​​the first obstacle in the structured road according to the first to-be-traveled area.

[0134] Optionally, the target to-be-traveled area determination subunit is specifically used to: if a crossable lane adjacent to the current lane is detected, then based on the guidance direction of the first driving guide line and the preset area length, determine the second to-be-traveled area located in front of the first obstacle in the crossable lane; based on the first to-be-traveled area and the second to-be-traveled area, determine the target to-be-traveled area of ​​the first obstacle in the structured road.

[0135] Optionally, the target future trajectory determination submodule includes:

[0136] A trajectory cost value determining unit, configured to determine a trajectory cost value corresponding to each future trajectory to be selected according to the first obstacle information and the second obstacle information;

[0137] The target future trajectory determination unit is used to determine the target future trajectory of the first obstacle according to the trajectory cost value corresponding to each candidate future trajectory.

[0138] Optionally, the trajectory cost value determining unit includes:

[0139] a current reference point cost value determining subunit, configured to determine a current reference point cost value corresponding to a current reference point to be driven in a selected future trajectory according to the first obstacle information, the second obstacle information and a previous reference point cost value corresponding to a previous reference point to be driven in a selected future trajectory;

[0140] The trajectory cost value determining subunit is used to determine the trajectory cost value corresponding to the future trajectory to be selected according to the reference point cost value corresponding to each reference point to be driven in the future trajectory to be selected.

[0141] Optionally, the current reference point cost value determination subunit is specifically used to:

[0142] Determine, according to the first obstacle information and the second obstacle information, a first generation value of the impact of the second obstacle on the first obstacle when the first obstacle travels from the last reference point to be traveled in the selected future trajectory to the current reference point to be traveled;

[0143] Determine a second generation value of the deviation of the first driving path according to the first driving guide line and the first driving path between the last reference point to be driven and the current reference point to be driven;

[0144] Determine a third generation value of the non-smoothness of the first driving path;

[0145] According to the previous reference point cost value, the first generation value, the second generation value and the third generation value corresponding to the previous reference point to be traveled, the current reference point cost value corresponding to the current reference point to be traveled in the future trajectory to be selected is determined.

[0146] Optionally, the current reference point cost value determination subunit is further specifically used for:

[0147] Determine, according to the first obstacle information and the second obstacle information, a collision cost value of a collision between the first obstacle and the first obstacle when the first obstacle travels from the last reference point to be traveled in the selected future trajectory to the current reference point to be traveled, and when the second obstacle travels from the first reference point to be traveled to the second reference point to be traveled, wherein the first reference point to be traveled is a reference point to be traveled corresponding to the second obstacle at the same layer as the last reference point to be traveled, and the second reference point to be traveled is a reference point to be traveled corresponding to the second obstacle at the same layer as the current reference point to be traveled;

[0148] Determine a deviation cost value of the second driving path according to a second driving guide line corresponding to the second obstacle and a second driving path between the first reference point to be driven and the second reference point to be driven;

[0149] Determining a non-smooth path cost value of the second driving path;

[0150] A first cost value of the impact of the second obstacle on the first obstacle is determined according to the collision cost value, the offset cost value and the path cost value.

[0151] The obstacle trajectory prediction device provided in the embodiment of the present invention can execute the obstacle trajectory prediction method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the obstacle trajectory prediction method.

[0152] It is worth noting that in the embodiment of the above-mentioned obstacle trajectory prediction device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0153] Embodiment 5

[0154] Figure 7 A schematic diagram of the structure of a device provided in Embodiment 5 of the present invention. Figure 7 A block diagram of an exemplary device 12 suitable for use in implementing embodiments of the present invention is shown. Figure 7 The device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0155] like Figure 7 As shown, device 12 is in the form of a general purpose computing device. Components of device 12 may include, but are not limited to, one or more processors or processing units 16, system memory 28, and bus 18 that connects various system components (including system memory 28 and processing unit 16).

[0156] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0157] Device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by device 12, including volatile and non-volatile media, removable and non-removable media.

[0158] System memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. Device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 7 not shown, usually called a "hard drive"). Although Figure 7 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The system memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0159] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0160] The device 12 may also communicate with one or more external devices 14 (e.g., keyboards, pointing devices, displays 24, etc.), may communicate with one or more devices that enable a user to interact with the device 12, and / or may communicate with any device that enables the device 12 to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed via an input / output (I / O) interface 22. Furthermore, the device 12 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 20. As shown, the network adapter 20 communicates with other modules of the device 12 via a bus 18. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the device 12, including, but not limited to, microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0161] The processing unit 16 executes various functional applications and data processing by running the program stored in the system memory 28, for example, implementing the steps of an obstacle trajectory prediction method provided by the embodiment of the present invention, the method comprising:

[0162] Acquire first obstacle information corresponding to a first obstacle traveling on a structured road and second obstacle information corresponding to a second obstacle within a preset range of the first obstacle;

[0163] Determining a first driving guide line corresponding to the first obstacle according to the road map information of the structured road and the first obstacle information;

[0164] A target future trajectory of the first obstacle is predicted according to the first driving guide line, the first obstacle information, and the second obstacle information.

[0165] Of course, those skilled in the art can understand that the processor can also implement the technical solution of the method for determining the reserved inventory amount provided in any embodiment of the present invention.

[0166] Embodiment 6

[0167] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the steps of the obstacle trajectory prediction method provided in any embodiment of the present invention are implemented. The method includes:

[0168] Acquire first obstacle information corresponding to a first obstacle traveling on a structured road and second obstacle information corresponding to a second obstacle within a preset range of the first obstacle;

[0169] Determining a first driving guide line corresponding to the first obstacle according to the road map information of the structured road and the first obstacle information;

[0170] A target future trajectory of the first obstacle is predicted according to the first driving guide line, the first obstacle information, and the second obstacle information.

[0171] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, 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 portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0172] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0173] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0174] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0175] It should be understood by those skilled in the art that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0176] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for predicting obstacle trajectories, characterized in that: include: Acquire first obstacle information corresponding to a first obstacle traveling on a structured road and second obstacle information corresponding to a second obstacle within a preset range of the first obstacle; Determining a first driving guide line corresponding to the first obstacle according to the road map information of the structured road and the first obstacle information; Predicting a target future trajectory of the first obstacle according to the first driving guide line, the first obstacle information, and the second obstacle information; The predicting of the target future trajectory of the first obstacle according to the first driving guide line, the first obstacle information and the second obstacle information includes: Determining, according to the first driving guide line, various candidate future trajectories of the first obstacle in the structured road; Determine, according to the first obstacle information and the second obstacle information, a trajectory cost value corresponding to each of the selected future trajectories; wherein the trajectory cost value is used to characterize the unreasonable degree of prediction of each of the selected future trajectories; A target future trajectory of the first obstacle is determined according to a trajectory cost value corresponding to each of the selected future trajectories.

2. The method according to claim 1, characterized in that Determining a first driving guide line corresponding to the first obstacle according to the road map information of the structured road and the first obstacle information includes: determining a current lane in which the first obstacle is located according to the road map information of the structured road and the first obstacle current position information of the first obstacle; A first driving guide line corresponding to the first obstacle is determined according to the current lane.

3. The method according to claim 2, characterized in that Determining, according to the current lane, a first driving guide line corresponding to the first obstacle, includes: Detecting whether a first driving reference direction of the current lane is consistent with a first obstacle driving direction of the first obstacle; If they are consistent, determining a first driving guide line corresponding to the first obstacle according to the first lane center line of the current lane and the first driving reference direction; If they are inconsistent, it is detected whether there is a crossable lane adjacent to the current lane, and the first driving guide line corresponding to the first obstacle is determined according to the detection result.

4. The method according to claim 3, characterized in that Determining a first driving guide line corresponding to the first obstacle according to the detection result includes: If there is a crossable lane adjacent to the current lane, determining a first driving guide line corresponding to the first obstacle according to a second driving reference direction of the crossable lane and a driving direction of the first obstacle; If there is no crossable lane adjacent to the current lane, a first driving guide line corresponding to the first obstacle is determined according to a first lane center line of the current lane and the first driving reference direction.

5. The method according to claim 4, characterized in that Determining a first driving guide line corresponding to the first obstacle according to the second driving reference direction that can cross the lane and the driving direction of the first obstacle includes: Detecting whether the second driving reference direction that can cross the lane is consistent with the first obstacle driving direction; If they are consistent, determining a first driving guide line corresponding to the first obstacle according to the second lane center line of the crossable lane and the second driving reference direction; If they are inconsistent, a first driving guide line corresponding to the first obstacle is determined according to the first lane center line of the current lane and the first driving reference direction.

6. The method according to claim 1, characterized in that Determining various candidate future trajectories of the first obstacle in the structured road according to the first driving guide line includes: Determining a target to-be-traveled area of ​​the first obstacle in the structured road according to the first driving guide line; The target area to be driven is divided into regions, and reference points to be driven on the boundary lines of each layer are determined; Determine various future trajectories to be selected corresponding to the first obstacle according to the reference point to be driven.

7. The method according to claim 6, characterized in that Determining a target to-be-traveled area of ​​the first obstacle in the structured road according to the first driving guide line includes: Based on the guiding direction of the first driving guide line and the preset area length, determining a first waiting area for driving located in front of the first obstacle in the current lane where the first obstacle is located; A target area to be driven of the first obstacle in the structured road is determined according to the first area to be driven.

8. The method according to claim 7, characterized in that Determining a target area to be driven by the first obstacle in the structured road according to the first area to be driven includes: If it is detected that there is a crossable lane adjacent to the current lane, determining a second waiting area for driving located in front of the first obstacle in the crossable lane based on the guidance direction of the first driving guide line and the preset area length; A target area to be driven of the first obstacle in the structured road is determined according to the first area to be driven and the second area to be driven.

9. The method according to claim 1, characterized in that: Determining a trajectory cost value corresponding to each of the selected future trajectories according to the first obstacle information and the second obstacle information includes: Determine a current reference point cost value corresponding to a current reference point to be driven in the selected future trajectory according to the first obstacle information, the second obstacle information, and a previous reference point cost value corresponding to a previous reference point to be driven in the selected future trajectory; According to the reference point cost value corresponding to each reference point to be driven in the future trajectory to be selected, a trajectory cost value corresponding to the future trajectory to be selected is determined.

10. The method according to claim 9, characterized in that Determining a current reference point cost value corresponding to a current reference point to be driven in the selected future trajectory according to the first obstacle information, the second obstacle information, and a previous reference point cost value corresponding to a previous reference point to be driven in the selected future trajectory includes: Determine, according to the first obstacle information and the second obstacle information, a first generation value of the impact of the second obstacle on the first obstacle when the first obstacle travels from the last reference point to be traveled in the selected future trajectory to the current reference point to be traveled; Determine a second generation value of the deviation of the first driving path according to the first driving guide line and a first driving path between the previous reference point to be driven and the current reference point to be driven; Determining a third generation value of the non-smoothness of the first driving path; Determine a current reference point cost value corresponding to a current reference point to be driven in the selected future trajectory according to the previous reference point cost value corresponding to the previous reference point to be driven, the first generation value, the second generation value, and the third generation value.

11. The method according to claim 10, characterized in that Determining, according to the first obstacle information and the second obstacle information, a first generation value of the impact of the second obstacle on the first obstacle when the first obstacle travels from the previous reference point to be traveled in the selected future trajectory to the current reference point to be traveled, includes: Determine, according to the first obstacle information and the second obstacle information, a collision cost of a collision between the first obstacle and the first obstacle when the first obstacle travels from a previous reference point to be traveled in the selected future trajectory to a current reference point to be traveled, and when the second obstacle travels from the first reference point to be traveled to a second reference point to be traveled, wherein the first reference point to be traveled is a reference point to be traveled corresponding to the second obstacle at the same layer as the previous reference point to be traveled, and the second reference point to be traveled is a reference point to be traveled corresponding to the second obstacle at the same layer as the current reference point to be traveled; Determine a deviation cost value of the second driving path according to a second driving guide line corresponding to the second obstacle and a second driving path between the first reference point to be driven and the second reference point to be driven; Determining a non-smooth path cost value of the second driving path; A first cost value of the impact of the second obstacle on the first obstacle is determined according to the collision cost value, the offset cost value, and the path cost value.

12. An obstacle trajectory prediction device, characterized in that: include: An obstacle information acquisition module, used to acquire first obstacle information corresponding to a first obstacle traveling on a structured road and second obstacle information corresponding to a second obstacle within a preset range of the first obstacle; A first driving guide line determining module, configured to determine a first driving guide line corresponding to the first obstacle according to the road map information of the structured road and the first obstacle information; a target future trajectory prediction module, configured to predict the target future trajectory of the first obstacle according to the first driving guide line, the first obstacle information and the second obstacle information; Wherein, the target future trajectory prediction module includes: A candidate future trajectory determination submodule, configured to determine each candidate future trajectory of the first obstacle in the structured road according to the first driving guide line; a target future trajectory determination submodule, configured to determine a target future trajectory of the first obstacle from the selected future trajectories according to the first obstacle information and the second obstacle information; The target future trajectory determination submodule includes: A trajectory cost value determining unit, configured to determine a trajectory cost value corresponding to each of the selected future trajectories according to the first obstacle information and the second obstacle information; wherein the trajectory cost value is used to characterize the unreasonable degree of prediction of each of the selected future trajectories; The target future trajectory determination unit is used to determine the target future trajectory of the first obstacle according to the trajectory cost value corresponding to each of the candidate future trajectories.

13. A device, characterized in that: The device comprises: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the obstacle trajectory prediction method as described in any one of claims 1-11.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the obstacle trajectory prediction method as described in any one of claims 1 to 11 is implemented.

Images