Vehicle surrounding obstacle trajectory prediction method, device, equipment and medium

Abstract

The present application relates to the technical field of trajectory prediction, and discloses a trajectory prediction method, device and equipment for a vehicle peripheral obstacle and a medium. When it is detected that an obstacle located at an intersection exists in the periphery of a vehicle, the present application calculates the lateral distance, the heading angle deviation, the relative steering and the steering score of the obstacle corresponding to each candidate driving lane by acquiring the multiple candidate driving lanes and the pose information of the obstacle. The driving intention of the obstacle is quantitatively analyzed by means of the steering score of the obstacle. Thus, the trajectory score of each candidate driving lane is obtained according to the lateral distance, the heading angle deviation, the relative steering and the steering score of the obstacle corresponding to each candidate driving lane. The driving trajectory of the obstacle is generated according to the trajectory score and the pose information of each candidate driving lane. The driving trajectory of the obstacle is predicted, which provides guidance for the path planning and obstacle avoidance of the ego vehicle, and prevents the unmanned vehicle from colliding with and rubbing against the obstacle.

Description

Technical Field

[0001] This invention relates to the field of trajectory prediction technology, specifically to methods, devices, equipment, and media for predicting the trajectory of obstacles around a vehicle. Background Technology

[0002] The core modules of an autonomous driving system include perception, prediction, planning and control, and high-precision positioning. Among these, the prediction module, based on environmental perception results, is responsible for predicting the trajectories of vehicles, pedestrians, and other obstacles over a future period, and then disseminating the prediction results to the decision-making or control modules. Therefore, the real-time performance and accuracy of the trajectory prediction results for surrounding obstacles directly determine the safety of autonomous vehicles and other road users.

[0003] Currently, trajectory prediction has relatively well-defined scenarios in structured road environments, such as lane-change trajectory prediction and intersection trajectory prediction, and high-precision map information is available as a reference for prediction in these specific scenarios. However, in park road scenarios, the road structure is complex, and high-precision maps accurate to the lane level cannot be provided. Furthermore, obstacles such as vehicles, electric vehicles, and pedestrians around the autonomous vehicle do not travel on structured roads, making obstacle trajectory prediction extremely complex.

[0004] Therefore, researching methods to accurately predict obstacle trajectories on park roads has great practical need and significance. Summary of the Invention

[0005] In view of this, the present invention provides a method, apparatus, device and medium for predicting the trajectory of obstacles around a vehicle, in order to solve the problem of being unable to accurately predict the trajectory of obstacles in complex road structures.

[0006] In a first aspect, the present invention provides a method for predicting the trajectory of obstacles around a vehicle, the method comprising:

[0007] When an obstacle located at an intersection is detected around the vehicle, multiple candidate driving lanes and pose information of the obstacle are obtained;

[0008] For each candidate driving lane, based on the trajectory and pose information of the current candidate driving lane, the lateral distance and heading angle deviation between the obstacle and the current candidate driving lane are obtained, and the relative steering of the current candidate driving lane relative to the obstacle is determined.

[0009] Based on the lateral distance corresponding to each candidate driving lane, the steering score of obstacles in each candidate driving lane is determined;

[0010] For each candidate driving lane, the trajectory score of the current candidate driving lane is determined based on the lateral distance, heading angle deviation, relative steering and obstacle steering scores corresponding to the current candidate driving lane.

[0011] The trajectory of the obstacle is obtained based on the trajectory score and pose information of each candidate driving lane.

[0012] By detecting obstacles at intersections around the vehicle, the system acquires multiple candidate lanes and pose information for the obstacles, calculates the lateral distance, heading angle deviation, relative steering, and obstacle steering score for each candidate lane, and quantifies the obstacle's driving intention based on the obstacle's steering score. This allows for the generation of trajectory scores for each candidate lane based on these parameters. Furthermore, by combining the trajectory scores and pose information of each candidate lane with the obstacle's trajectory, the system predicts the obstacle's path, providing guidance for autonomous driving path planning and obstacle avoidance, and preventing collisions and friction between the autonomous vehicle and obstacles.

[0013] In one optional implementation, the lateral distance includes lateral distances at multiple times; the trajectory score of the current candidate driving lane is determined based on the lateral distance, heading angle deviation, relative steering, and obstacle steering score corresponding to the current candidate driving lane, including:

[0014] The lateral distance variance is obtained based on the lateral distances at multiple times corresponding to the current candidate driving lane.

[0015] Based on the steering scores relative to the steering and obstacles, the target steering score and steering matching score of the current candidate driving lane are obtained;

[0016] The trajectory score of the current candidate driving lane is obtained based on the lateral distance variance, heading angle deviation, target steering score, and steering matching score.

[0017] This allows for the accurate calculation of the trajectory score of the current candidate driving lane using lateral distance variance, heading angle deviation, target steering score, and steering matching score. This quantitative evaluation of the candidate driving lane provides guidance for selecting the possible target driving lane for obstacles.

[0018] In one optional implementation, the trajectory score of the current candidate driving lane is obtained based on the lateral distance variance, heading angle deviation, target steering score, and steering matching score, including:

[0019] Calculate the product of the heading angle deviation and the preset ratio, and sum the product, the lateral distance variance, the target steering score, and the negative of the steering matching score to obtain the trajectory score of the current candidate driving lane.

[0020] In one optional implementation, the target steering score and steering matching score of the current candidate driving lane are obtained based on the steering scores relative to the steering and obstacles, including:

[0021] The steering score of obstacles is updated based on relative steering to obtain the target steering score of the current candidate driving lane; relative steering includes going straight, turning left, and turning right.

[0022] Based on the obstacle's steering score, determine the obstacle's driving intention; the driving intention includes going straight, turning left, and turning right;

[0023] The driving intention of the obstacle is compared with the relative steering, and the steering matching score of the current candidate driving lane is calculated based on the comparison result and the target steering score.

[0024] Based on the relative steering and obstacle steering scores, the target steering score and steering matching score of the current candidate driving lane are obtained. The candidate driving lane is evaluated from several aspects, including the driving intention of the obstacle, the relative steering of the lane, and the degree of matching between the relative steering of the lane and the driving intention of the obstacle.

[0025] In one optional implementation, the trajectory of the obstacle is obtained based on the trajectory score and pose information of each candidate driving lane, including:

[0026] Based on the trajectory scores of each candidate driving lane, the candidate driving lane with the lowest trajectory score is determined as the target driving lane for the obstacle.

[0027] Based on the target driving lane and the position and heading angle of the obstacle in the pose information, the driving trajectory of the obstacle is generated.

[0028] By identifying the candidate driving lane with the lowest trajectory score as the target driving lane for the obstacle, and generating the obstacle's driving trajectory based on the target driving lane and the obstacle's position and heading angle in the pose information, the driving trajectory of the obstacle can be predicted for a period of time in the future, providing reference and guidance for path planning and obstacle avoidance of autonomous vehicles.

[0029] In one optional implementation, the pose information includes the pose information of the obstacle at multiple times; for each candidate driving lane, based on the trajectory information and pose information of the current candidate driving lane, the lateral distance and heading angle deviation between the obstacle and the current candidate driving lane are obtained, and the relative steering of the current candidate driving lane relative to the obstacle is determined, including:

[0030] For each candidate driving lane, the lateral distance between the obstacle and the current candidate driving lane is obtained at each time step based on the pose information of the obstacle at multiple time steps and the trajectory information of the current candidate driving lane.

[0031] By using the pose information of the obstacle at multiple moments and the trajectory information of the current candidate driving lane, the lateral distance between the obstacle and the current candidate driving lane at each moment can be obtained, so as to analyze the positional changes between the obstacle and the lane.

[0032] In one optional implementation, the steering score of obstacles in each candidate driving lane is determined based on the lateral distance corresponding to each candidate driving lane, including:

[0033] For each candidate driving lane, based on the lateral distance of the current candidate driving lane at multiple times, iterate through and calculate the difference between the lateral distance at the last time and the lateral distance at each time, and calculate the turning score of the obstacle in the current candidate driving lane based on the difference.

[0034] By measuring the lateral distance between the obstacle and the candidate driving lane at multiple time points, the steering score of the obstacle in the candidate driving lane can be obtained, so as to analyze the steering intention of the obstacle in subsequent steps.

[0035] In a second aspect, the present invention provides a trajectory prediction device for obstacles around a vehicle, the device comprising:

[0036] The acquisition module is used to acquire multiple candidate driving lanes and pose information of obstacles located at intersections when obstacles are detected around the vehicle.

[0037] The first processing module is used to obtain the lateral distance and heading angle deviation between the obstacle and the current candidate driving lane based on the trajectory information and pose information of the current candidate driving lane for each candidate driving lane, and to determine the relative steering of the current candidate driving lane relative to the obstacle.

[0038] The second processing module is used to determine the steering score of obstacles in each candidate driving lane based on the lateral distance corresponding to each candidate driving lane.

[0039] The third processing module is used to determine the trajectory score of each candidate driving lane based on the lateral distance, heading angle deviation, relative steering and obstacle steering score corresponding to the current candidate driving lane.

[0040] The fourth processing module is used to obtain the trajectory of the obstacle based on the trajectory score and pose information of each candidate driving lane.

[0041] Thirdly, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the trajectory prediction method for obstacles around a vehicle as described in the first aspect or any corresponding embodiment thereof.

[0042] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the trajectory prediction method for obstacles around a vehicle according to the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0043] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0044] Figure 1 This is a flowchart illustrating a method for predicting the trajectory of obstacles around a vehicle according to an embodiment of the present invention.

[0045] Figure 2 This is a schematic diagram of an intersection area according to an embodiment of the present invention;

[0046] Figure 3 This is a flowchart illustrating another method for predicting the trajectory of obstacles around a vehicle according to an embodiment of the present invention;

[0047] Figure 4 This is a schematic diagram showing the positions of obstacles and candidate driving lanes according to an embodiment of the present invention;

[0048] Figure 5 This is a schematic diagram illustrating the application of the trajectory prediction method for obstacles around a vehicle according to an embodiment of the present invention;

[0049] Figure 6 This is a structural block diagram of a trajectory prediction device for obstacles around a vehicle according to an embodiment of the present invention;

[0050] Figure 7 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] According to an embodiment of the present invention, a method for predicting the trajectory of obstacles around a vehicle is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0053] This embodiment provides a method for predicting the trajectory of obstacles around a vehicle, which can be used with computer or electronic equipment, such as an onboard computer or vehicle controller, to predict the trajectory of obstacles around a vehicle. Figure 1 This is a flowchart of a method for predicting the trajectory of obstacles around a vehicle according to an embodiment of the present invention, as shown below. Figure 1 As shown, the process includes the following steps:

[0054] Step S101: When an obstacle located at an intersection is detected around the vehicle, multiple candidate driving lanes and pose information of the obstacle are obtained.

[0055] Specifically, data will be collected during map creation, such as... Figure 2 The intersection area shown is a polygonal region. When an autonomous vehicle detects an obstacle in the vicinity of the intersection, it acquires the obstacle's pose information over a period of time (including its position and heading angle at multiple consecutive moments) and the candidate lanes the obstacle might take. This information is used to predict the obstacle's trajectory over a future period, providing data guidance and reference for the vehicle's path planning and preventing collisions and friction with obstacles around the vehicle. It should be noted that obstacles can be moving objects such as vehicles, bicycles, and pedestrians.

[0056] Step S102: For each candidate driving lane, based on the trajectory information and pose information of the current candidate driving lane, obtain the lateral distance and heading angle deviation between the obstacle and the current candidate driving lane, and determine the relative steering of the current candidate driving lane relative to the obstacle.

[0057] Specifically, the lateral distance between the obstacle and the current candidate driving lane refers to the vertical lateral distance between the obstacle and the centerline of the current candidate driving lane. Based on the position of the obstacle at multiple consecutive moments contained in the obstacle pose information, the lateral distance of the obstacle at each of the multiple consecutive moments can be calculated, thereby analyzing the driving intention of the obstacle and providing a basis for the selection of candidate lanes.

[0058] Specifically, the relative direction (straight, left turn, right turn) of a lane can be determined based on its relative position to the obstacle. The two points closest to the vehicle from the trajectory points of the candidate driving lane, and the last point along the vehicle's direction of travel from the trajectory points of the candidate driving lane, are selected. If the error in the orientation angle between the two closest points and the last point is less than a certain angle, the relative direction is considered to be straight. Otherwise, the left or right orientation is determined based on the vector relationship between the three points. For details, please refer to the description of the relevant technology; it will not be elaborated upon here.

[0059] Specifically, the heading angle deviation α between the obstacle and the current candidate driving lane can be calculated using the following formula:

[0060] α=θ ve h-lane

[0061] Where, θ veh θ represents the heading angle of the obstacle. lane Indicates the heading angle of the lane.

[0062] Step S103: Determine the steering score of obstacles in each candidate driving lane based on the lateral distance corresponding to each candidate driving lane.

[0063] Specifically, the steering score of an obstacle in a candidate driving lane can be calculated by measuring the lateral distance of the obstacle at multiple consecutive moments, thereby quantifying the obstacle's driving intention and providing a basis for the selection of candidate lanes.

[0064] Step S104: For each candidate driving lane, determine the trajectory score of the current candidate driving lane based on the lateral distance, heading angle deviation, relative steering, and obstacle steering scores corresponding to the current candidate driving lane.

[0065] Specifically, the trajectory score of the candidate driving lane is calculated from several aspects, including lateral distance, heading angle deviation, relative lane steering, and the driving intention of obstacles, so as to evaluate the candidate driving lane.

[0066] Step S105: Obtain the trajectory of the obstacle based on the trajectory score and pose information of each candidate driving lane.

[0067] Specifically, after obtaining the trajectory scores of each candidate driving lane, the lane that the obstacle is most likely to travel in in the future can be determined. Then, based on the obstacle's current position and heading angle, the obstacle's driving trajectory can be generated along the trajectory position of the selected driving lane, thereby providing guidance for the autonomous driving path planning and obstacle avoidance of the vehicle.

[0068] The trajectory prediction method for obstacles around a vehicle provided in this embodiment, when an obstacle located at an intersection is detected around the vehicle, obtains multiple candidate driving lanes and pose information of the obstacle, calculates the lateral distance, heading angle deviation, relative steering, and steering score of each candidate driving lane, and performs a quantitative analysis of the obstacle's driving intention based on the obstacle's steering score, thereby obtaining the trajectory score of each candidate driving lane, and generating the driving trajectory of the obstacle based on the trajectory score and pose information of each candidate driving lane, thus realizing the prediction of the obstacle's driving trajectory and providing guidance for autonomous driving path planning and obstacle avoidance of the vehicle.

[0069] This embodiment provides a method for predicting the trajectory of obstacles around a vehicle, which can be used with computer or electronic equipment, such as an onboard computer or vehicle controller, to predict the trajectory of obstacles around a vehicle. Figure 3 This is a flowchart of a method for predicting the trajectory of obstacles around a vehicle according to an embodiment of the present invention, as shown below. Figure 3 As shown, the process includes the following steps:

[0070] Step S301: When an obstacle located at an intersection is detected around the vehicle, acquire multiple candidate driving lanes and pose information of the obstacle. For details, please refer to [link to relevant documentation]. Figure 1 Step S101 of the illustrated embodiment will not be described again here.

[0071] Step S302: For each candidate driving lane, based on the trajectory information and pose information of the current candidate driving lane, obtain the lateral distance and heading angle deviation between the obstacle and the current candidate driving lane, and determine the relative steering of the current candidate driving lane relative to the obstacle.

[0072] Specifically, the pose information includes the pose information of the obstacle at multiple times. For each candidate driving lane, based on the pose information of the obstacle at multiple times and the trajectory information of the current candidate driving lane, the lateral distance between the obstacle and the current candidate driving lane at each time is obtained.

[0073] In some alternative implementations, such as Figure 4 As shown, let the position of the obstacle at a certain moment be P(x0,y0). Select the two points closest to the obstacle on the lane line of the current candidate driving lane, and denote their positions as A(x1,y1) and B(x2,y2). Then, the following relationship holds:

[0074]

[0075] The lateral distance d between the obstacle and the current candidate driving lane can be calculated using the following formula:

[0076]

[0077] By using the pose information of the obstacle at multiple moments and the trajectory information of the current candidate driving lane, the lateral distance between the obstacle and the current candidate driving lane at each moment can be obtained, so as to analyze the positional changes between the obstacle and the lane.

[0078] Step S303: Determine the steering score of obstacles in each candidate driving lane based on the lateral distance corresponding to each candidate driving lane.

[0079] Specifically, for each candidate driving lane, based on the lateral distance of the current candidate driving lane at multiple times, the difference between the lateral distance at the last time and the lateral distance at each time is calculated, and the turning score of the obstacle in the current candidate driving lane is calculated based on the difference.

[0080] In some optional implementations, the lateral distance at the last moment is denoted as dl. The lateral distances di corresponding to multiple moments within a certain time period (e.g., within 2 seconds) are iterated. When the difference between dl and di is greater than a certain threshold (which can be set according to actual application requirements, e.g., 0.2m), the turning score is increased by ki*(dl-di). The turning score is obtained by iterating through the lateral distances corresponding to all moments, where ki represents the scaling factor, which can be set according to the actual scenario.

[0081] By measuring the lateral distance between the obstacle and the candidate driving lane at multiple time points, the steering score of the obstacle in the candidate driving lane can be obtained, so as to analyze the steering intention of the obstacle in subsequent steps.

[0082] Step S304: For each candidate driving lane, determine the trajectory score of the current candidate driving lane based on the lateral distance, heading angle deviation, relative steering, and obstacle steering scores corresponding to the current candidate driving lane.

[0083] Specifically, the lateral distance includes the lateral distance at multiple times, and step S204 above includes:

[0084] Step S3041: Obtain the lateral distance variance value based on the lateral distances at multiple times corresponding to the current candidate driving lane.

[0085] Specifically, the average lateral distance at multiple moments within a certain time period (e.g., within 2 seconds) can be calculated using the following formula.

[0086]

[0087] Where n represents the number of time points.

[0088] Next, based on the average value of the lateral distance d at multiple moments within a certain time period... The variance of the lateral distance σ is calculated using the following formula. 2 :

[0089]

[0090] Step S3042: Based on the relative steering and obstacle steering scores, obtain the target steering score and steering matching score of the current candidate driving lane.

[0091] In some optional implementations, step S3042 above includes:

[0092] Step a1: Update the steering score of the obstacle based on relative steering to obtain the target steering score of the current candidate driving lane, where relative steering includes going straight, turning left and turning right.

[0093] Specifically, if the relative turn is left, the target turn score is obtained by inverting the obstacle's turn score; if the relative turn is straight, the target turn score is directly obtained by taking the obstacle's turn score; if the relative turn is right, the target turn score is obtained by taking the absolute value of the obstacle's turn score.

[0094] Step a2: Determine the driving intention of the obstacle based on its steering score, where the driving intention includes going straight, turning left, and turning right.

[0095] Specifically, the obstacle's steering score is compared with a preset threshold. If the steering score is less than a preset left-turn threshold, the obstacle's driving intention is to turn left; if the steering score is greater than a preset right-turn threshold, the obstacle's driving intention is to turn right; otherwise, the obstacle's driving intention is to go straight. For example, the preset left-turn threshold can be -0.02, and the preset right-turn threshold can be 0.02. In practice, these can be set according to specific application scenarios, and this invention is not limited thereto.

[0096] Step a3: Compare the driving intention of the obstacle with the relative steering, and calculate the steering matching score of the current candidate driving lane based on the comparison result and the target steering score.

[0097] Specifically, if the obstacle's driving intention does not match the relative steering, for example, if the obstacle's driving intention is to turn left while the relative steering is to turn right, then the target steering score is reduced by a certain percentage to obtain a steering matching score. For example, the steering matching score is the target steering score / 2.

[0098] Based on the relative steering and obstacle steering scores, the target steering score and steering matching score of the current candidate driving lane are obtained. The candidate driving lane is evaluated from several aspects, including the driving intention of the obstacle, the relative steering of the lane, and the degree of matching between the relative steering of the lane and the driving intention of the obstacle.

[0099] Step S3043: Based on the lateral distance variance, heading angle deviation, target steering score, and steering matching score, obtain the trajectory score of the current candidate driving lane.

[0100] Specifically, the product of the heading angle deviation and the preset ratio is calculated, and the product, the lateral distance variance, the target steering score, and the negative of the steering matching score are summed to obtain the trajectory score of the current candidate driving lane.

[0101] In some alternative implementations, the trajectory score S of the current candidate driving lane can be calculated according to the following formula:

[0102] S=k*α+σ 2 +DV

[0103] Where k represents a preset ratio, D represents the target steering score, and V represents the steering matching score. For example, k can be 0.3, but the present invention is not limited thereto.

[0104] This allows for the accurate calculation of the trajectory score of the current candidate driving lane using lateral distance variance, heading angle deviation, target steering score, and steering matching score. This quantitative evaluation of the candidate driving lane provides guidance for selecting the possible target driving lane for obstacles.

[0105] Step S305: Obtain the trajectory of the obstacle based on the trajectory score and pose information of each candidate driving lane.

[0106] Specifically, step S305 includes:

[0107] Step S3051: Based on the trajectory scores of each candidate driving lane, the candidate driving lane with the smallest trajectory score is determined as the target driving lane of the obstacle.

[0108] Step S3052: Based on the target driving lane and the position and heading angle of the obstacle in the pose information, generate the driving trajectory of the obstacle.

[0109] Specifically, after determining the most likely target lane for the obstacle, the obstacle's trajectory is generated from its current position along the heading angle and the location of the trajectory point in the target lane. It is important to note that the lateral distance between this trajectory and the obstacle is the same as the lateral distance between the obstacle and the target lane.

[0110] By identifying the candidate driving lane with the lowest trajectory score as the target driving lane for the obstacle, and generating the obstacle's driving trajectory based on the target driving lane and the obstacle's position and heading angle in the pose information, the driving trajectory of the obstacle can be predicted for a period of time in the future, providing reference and guidance for path planning and obstacle avoidance of autonomous vehicles.

[0111] The trajectory prediction method for obstacles around a vehicle provided in this embodiment, when an obstacle located at an intersection is detected around the vehicle, calculates the lateral distance, heading angle deviation, relative steering, and steering score of each candidate driving lane by acquiring multiple candidate driving lanes and pose information of the obstacle. The driving intention of the obstacle is quantitatively analyzed based on the steering score of the obstacle, thereby obtaining the trajectory score of each candidate driving lane. The candidate driving lane with the smallest trajectory score is determined as the target driving lane of the obstacle. Based on the target driving lane and the position and heading angle of the obstacle in the pose information, the driving trajectory of the obstacle is generated, realizing the prediction of the driving trajectory of the obstacle for a period of time in the future. This provides guidance for the autonomous driving path planning and obstacle avoidance of the vehicle, avoiding collisions and friction with obstacles.

[0112] The trajectory prediction method for obstacles around a vehicle according to the present invention will be described in detail below with reference to a specific application example, such as... Figure 5 As shown, this specific application example includes the following steps:

[0113] 1. Obtain information such as the location, speed, and heading angle of the obstacle, as well as map information of the intersection area where the obstacle is located. This map information is used to pre-collect the lane lines that the obstacle may travel on.

[0114] 2. Calculate the lateral distance and heading angle error between the obstacle and the pre-collected lane line, and calculate the average lateral distance and lateral distance variance based on the lateral distance over a certain period of time.

[0115] 3. Based on the calculated lateral distance and heading angle error between the obstacle and the pre-collected lane line, determine the relative steering of the pre-collected lane line relative to the obstacle and the driving intention of the obstacle.

[0116] 4. Based on the lateral distance variance, heading angle error, relative steering of the pre-collected lane line to the obstacle, and the obstacle's driving intention, calculate the trajectory score of the pre-collected lane line, select the lane line with the smallest trajectory score as the lane line that the obstacle is most likely to choose to drive on, and thus predict the obstacle's trajectory.

[0117] This embodiment also provides a trajectory prediction device for obstacles around a vehicle. This device is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0118] This embodiment provides a trajectory prediction device for obstacles around a vehicle, such as... Figure 6 As shown, it includes:

[0119] The acquisition module 601 is used to acquire multiple candidate driving lanes and pose information of the obstacle when an obstacle located at an intersection is detected around the vehicle.

[0120] The first processing module 602 is used to obtain the lateral distance and heading angle deviation between the obstacle and the current candidate driving lane based on the trajectory information and pose information of the current candidate driving lane for each candidate driving lane, and to determine the relative steering of the current candidate driving lane relative to the obstacle.

[0121] The second processing module 603 is used to determine the steering score of obstacles in each candidate driving lane based on the lateral distance corresponding to each candidate driving lane.

[0122] The third processing module 604 is used to determine the trajectory score of each candidate driving lane based on the lateral distance, heading angle deviation, relative steering and obstacle steering score corresponding to the current candidate driving lane.

[0123] The fourth processing module 605 is used to obtain the driving trajectory of the obstacle based on the trajectory score and pose information of each candidate driving lane.

[0124] In some optional implementations, the pose information includes the pose information of the obstacle at multiple moments; the first processing module 602 includes:

[0125] The first processing unit is used to obtain the lateral distance between the obstacle and the current candidate driving lane at each time step, based on the pose information of the obstacle at multiple times and the trajectory information of the current candidate driving lane.

[0126] In some alternative implementations, the second processing module 603 includes:

[0127] The second processing unit is used to calculate the difference between the lateral distance at the last moment and the lateral distance at each moment for each candidate driving lane, based on the lateral distance of the current candidate driving lane at multiple moments, and calculate the turning score of the obstacle in the current candidate driving lane based on the difference.

[0128] In some optional implementations, the lateral distance includes lateral distances at multiple times; the third processing module 604 includes:

[0129] The third processing unit is used to obtain the lateral distance variance value based on the lateral distances at multiple times corresponding to the current candidate driving lane;

[0130] The fourth processing unit is used to obtain the target steering score and steering matching score of the current candidate driving lane based on the steering scores of relative steering and obstacles;

[0131] The fifth processing unit is used to obtain the trajectory score of the current candidate driving lane based on the lateral distance variance, heading angle deviation, target steering score, and steering matching score.

[0132] In some alternative implementations, the fourth processing unit includes:

[0133] The first processing subunit is used to update the steering score of obstacles based on relative steering to obtain the target steering score of the current candidate driving lane; relative steering includes going straight, turning left, and turning right;

[0134] The second processing subunit is used to determine the obstacle's driving intention based on the obstacle's steering score; the driving intention includes going straight, turning left, and turning right.

[0135] The third processing subunit is used to compare the driving intention of the obstacle with the relative steering, and calculate the steering matching score of the current candidate driving lane based on the comparison result and the target steering score.

[0136] In some alternative implementations, the fifth processing unit includes:

[0137] The fourth processing subunit is used to calculate the product of the heading angle deviation and the preset ratio, and sum the product, the lateral distance variance, the target steering score, and the negative of the steering matching score to obtain the trajectory score of the current candidate driving lane.

[0138] In some alternative implementations, the fourth processing module 605 includes:

[0139] The sixth processing unit is used to determine the candidate driving lane with the smallest trajectory score as the target driving lane of the obstacle based on the trajectory score of each candidate driving lane.

[0140] The seventh processing unit is used to generate the driving trajectory of the obstacle based on the target driving lane and the position and heading angle of the obstacle in the pose information.

[0141] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0142] In this embodiment, the trajectory prediction device for obstacles around the vehicle is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0143] This invention also provides a computer device having the above-described features. Figure 6 The device shown is a trajectory prediction device for obstacles around a vehicle.

[0144] Please see Figure 7 , Figure 7 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 7 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 7 Take a processor 10 as an example.

[0145] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0146] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0147] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0148] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0149] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0150] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0151] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for predicting the trajectory of obstacles around a vehicle, characterized in that, The method includes: When an obstacle located at an intersection is detected around the vehicle, multiple candidate driving lanes and position information of the obstacle are obtained; For each candidate driving lane, based on the trajectory information of the current candidate driving lane and the pose information of the obstacle, the lateral distance and heading angle deviation between the obstacle and the current candidate driving lane are obtained, and the relative steering of the current candidate driving lane relative to the obstacle is determined. Based on the lateral distance corresponding to each candidate driving lane, the steering score of the obstacle in each candidate driving lane is determined; For each candidate driving lane, the trajectory score of the current candidate driving lane is determined based on the lateral distance, heading angle deviation, relative steering, and steering score of the obstacle corresponding to the current candidate driving lane. The driving trajectory of the obstacle is obtained based on the trajectory score of each candidate driving lane and the pose information of the obstacle.

2. The method according to claim 1, characterized in that, The lateral distance includes lateral distances at multiple times; determining the trajectory score of the current candidate driving lane based on the lateral distance, heading angle deviation, relative steering, and steering score of the obstacle, includes: The lateral distance variance is obtained based on the lateral distances at multiple times corresponding to the current candidate driving lane. Based on the relative steering and the steering scores of the obstacles, the target steering score and steering matching score of the current candidate driving lane are obtained; The trajectory score of the current candidate driving lane is obtained based on the lateral distance variance, the heading angle deviation, the target steering score, and the steering matching score.

3. The method according to claim 2, characterized in that, The process of obtaining the trajectory score of the current candidate driving lane based on the lateral distance variance, the heading angle deviation, the target steering score, and the steering matching score includes: Calculate the product of the heading angle deviation and the preset ratio, and sum the product, the lateral distance variance, the target steering score, and the negative of the steering matching score to obtain the trajectory score of the current candidate driving lane.

4. The method according to claim 2, characterized in that, The step of obtaining the target steering score and steering matching score of the current candidate driving lane based on the relative steering and the steering score of the obstacle includes: The steering score of the obstacle is updated based on the relative steering to obtain the target steering score of the current candidate driving lane; the relative steering includes going straight, turning left, and turning right. Based on the obstacle's steering score, the obstacle's driving intention is determined; the driving intention includes going straight, turning left, and turning right. The driving intention of the obstacle is compared with the relative steering, and the steering matching score of the current candidate driving lane is calculated based on the comparison result and the target steering score.

5. The method according to claim 1, characterized in that, The step of obtaining the obstacle's trajectory based on the trajectory scores of each candidate driving lane and the obstacle's pose information includes: Based on the trajectory scores of each candidate driving lane, the candidate driving lane with the lowest trajectory score is determined as the target driving lane of the obstacle. Based on the target driving lane and the position and heading angle of the obstacle in the pose information, the driving trajectory of the obstacle is generated.

6. The method according to claim 1, characterized in that, The pose information includes the pose information of the obstacle at multiple moments; for each candidate driving lane, based on the trajectory information of the current candidate driving lane and the pose information of the obstacle, the lateral distance and heading angle deviation between the obstacle and the current candidate driving lane are obtained, and the relative steering of the current candidate driving lane relative to the obstacle is determined, including: For each candidate driving lane, based on the pose information of the obstacle at multiple times and the trajectory information of the current candidate driving lane, the lateral distance between the obstacle and the current candidate driving lane at each time moment is obtained.

7. The method according to claim 2 or 5, characterized in that, The step of determining the steering score of the obstacle in each candidate driving lane based on the lateral distance corresponding to each candidate driving lane includes: For each candidate driving lane, based on the lateral distance of the current candidate driving lane at multiple times, the difference between the lateral distance at the last time and the lateral distance at each time is calculated, and the turning score of the obstacle in the current candidate driving lane is calculated based on the difference.

8. A trajectory prediction device for obstacles around a vehicle, characterized in that, The device includes: The acquisition module is used to acquire multiple candidate driving lanes and pose information of the obstacle when an obstacle located at an intersection is detected around the vehicle; The first processing module is used to, for each candidate driving lane, obtain the lateral distance and heading angle deviation between the obstacle and the current candidate driving lane based on the trajectory information of the current candidate driving lane and the pose information of the obstacle, and determine the relative steering of the current candidate driving lane relative to the obstacle; The second processing module is used to determine the steering score of the obstacle in each candidate driving lane based on the lateral distance corresponding to each candidate driving lane. The third processing module is used to determine the trajectory score of each candidate driving lane based on the lateral distance, heading angle deviation, relative steering, and steering score of the obstacle. The fourth processing module is used to obtain the driving trajectory of the obstacle based on the trajectory score of each candidate driving lane and the pose information of the obstacle.

9. A computer device, characterized in that, include: A memory and a processor are communicatively connected, the memory storing computer instructions, and the processor executing the computer instructions to perform the trajectory prediction method for obstacles around a vehicle as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the trajectory prediction method for obstacles around the vehicle as described in any one of claims 1 to 7.

Images