Obstacle orientation acquisition method and device, electronic equipment and storage medium

Abstract

The invention provides an obstacle orientation acquisition method and device, electronic equipment and a storage medium, and relates to the field of artificial intelligence such as automatic driving, computer vision and deep learning. The method can comprise the steps that for any obstacle detected in the vehicle driving process, the predicted moving track of the obstacle in the future preset duration is obtained, the moving track is composed of a starting track point and a non-starting track point, and the starting track point is the current position point of the obstacle; determining a target orientation of the obstacle at the starting track point according to the obtained predicted orientation of the obstacle at the starting track point and the speed direction; and for each non-initial trajectory point, determining the target orientation of the obstacle at the non-initial trajectory point according to the obtained speed direction of the obstacle at the non-initial trajectory point, the correction coefficient and the initial correction amount corresponding to the non-initial trajectory point. By applying the scheme of the invention, the accuracy and the like of the determined orientation can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to methods, devices, electronic devices and storage media for obtaining obstacle orientations in the fields of autonomous driving, computer vision and deep learning. Background Art

[0002] In the autonomous driving scenario, obstacles can be detected in real time while the vehicle is driving, and the deep learning model can be used to predict the movement trajectory of any obstacle detected within a predetermined time period in the future. The movement trajectory consists of multiple trajectory points. Generally speaking, the deep learning model only outputs the coordinate information of each trajectory point, but the downstream planning module needs to use both the direction and coordinate information of the obstacle when projecting the obstacle for trajectory planning. Summary of the invention

[0003] The present disclosure provides a method, device, electronic device and storage medium for obtaining an obstacle direction.

[0004] A method for obtaining an obstacle direction, comprising:

[0005] For any obstacle detected during vehicle driving, obtain a predicted moving trajectory of the obstacle within a predetermined time period in the future, wherein the moving trajectory consists of a starting trajectory point and non-starting trajectory points, and the starting trajectory point is the current location of the obstacle;

[0006] Acquire the speed direction of the obstacle at the starting trajectory point as a first reference direction, and acquire the predicted orientation of the obstacle at the starting trajectory point, and determine the target orientation of the obstacle at the starting trajectory point according to a first deviation angle between the predicted orientation and the first reference direction, wherein the target orientation is the orientation determined as a desired result;

[0007] For each non-starting trajectory point, the speed direction of the obstacle at the non-starting trajectory point is obtained as the second reference direction, and an initial correction amount corresponding to the non-starting trajectory point is determined; the target direction of the obstacle at the non-starting trajectory point is determined according to the initial correction amount, the correction coefficient and the second reference direction, wherein the correction coefficient is determined according to the target direction at the starting trajectory point.

[0008] An obstacle direction acquisition device includes: a trajectory acquisition module, a first determination module and a second determination module;

[0009] The trajectory acquisition module is used to acquire the predicted moving trajectory of any obstacle detected during the driving process of the vehicle within a predetermined time period in the future, wherein the moving trajectory consists of a starting trajectory point and non-starting trajectory points, and the starting trajectory point is the current location of the obstacle;

[0010] The first determination module is used to obtain the speed direction of the obstacle at the starting trajectory point as a first reference direction, and obtain the predicted orientation of the obstacle at the starting trajectory point, and determine the target orientation of the obstacle at the starting trajectory point according to a first deviation angle between the predicted orientation and the first reference direction, wherein the target orientation is the orientation determined as the desired result;

[0011] The second determination module is used to obtain the speed direction of the obstacle at each non-starting trajectory point as the second reference direction, and determine the initial correction amount corresponding to the non-starting trajectory point, and determine the target direction of the obstacle at the non-starting trajectory point according to the initial correction amount, the correction coefficient and the second reference direction, wherein the correction coefficient is determined according to the target direction at the starting trajectory point.

[0012] An electronic device, comprising:

[0013] at least one processor; and

[0014] a memory communicatively connected to the at least one processor; wherein,

[0015] The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the method described above.

[0016] A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method as described above.

[0017] A computer program product comprises a computer program / instruction, wherein the computer program / instruction implements the method described above when executed by a processor.

[0018] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0020] Figure 1 It is a flow chart of an embodiment of the obstacle orientation acquisition method disclosed in the present invention;

[0021] Figure 2 A schematic diagram of the relationship between the vehicle body orientation and the speed direction during the vehicle turning process described in the present disclosure;

[0022] Figure 3 A schematic diagram of the relationship between the angle between the vehicle body direction and the speed direction and the turning curvature radius during the turning process of the vehicle disclosed in the present invention;

[0023] Figure 4 It is a schematic diagram of the structure of an obstacle direction acquisition device embodiment 400 of the present disclosure;

[0024] Figure 5 A schematic block diagram of an electronic device 500 that can be used to implement an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0025] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0026] In addition, it should be understood that the term "and / or" in this article is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0027] Figure 1 FIG. 1 is a flow chart of an embodiment of the obstacle orientation acquisition method described in the present disclosure. Figure 1 As shown, the following specific implementation methods are included.

[0028] In step 101, for any obstacle detected during vehicle driving, a predicted moving trajectory of the obstacle within a predetermined time period in the future is obtained, wherein the moving trajectory consists of a starting trajectory point and non-starting trajectory points, and the starting trajectory point is the current location of the obstacle.

[0029] In step 102, the speed direction of the obstacle at the starting trajectory point is obtained as the first reference direction, and the predicted orientation of the obstacle at the starting trajectory point is obtained, and the target orientation of the obstacle at the starting trajectory point is determined based on a first deviation angle between the predicted orientation and the first reference direction, and the target orientation is the orientation determined as the desired result.

[0030] In step 103, for each non-starting trajectory point, the speed direction of the obstacle at the non-starting trajectory point is obtained as the second reference direction, and the initial correction amount corresponding to the non-starting trajectory point is determined. The target direction of the obstacle at the non-starting trajectory point is determined according to the initial correction amount, the correction coefficient and the second reference direction, wherein the correction coefficient is determined according to the target direction at the starting trajectory point.

[0031] The obstacles described in the present disclosure are generally dynamic obstacles, so there will be moving tracks.

[0032] In the traditional method, the speed direction (i.e., displacement direction) of the obstacle at each trajectory point is usually directly used as the direction of the obstacle at each trajectory point. However, in many cases, there will be certain differences between the speed direction and the direction of the obstacle. For example, for obstacles such as motor vehicles, due to the long body, there will be certain differences between the speed direction and the direction of the obstacle (i.e., the direction of the vehicle body) during the turning process. Especially when the turning curvature is large, the difference will be particularly obvious. Accordingly, if the speed direction is used as the direction of the obstacle, it may cause errors in the projection during planning, which may lead to unreasonable planning trajectories, which may cause collision risks or dangerous behaviors such as sudden braking, reducing the safety of vehicle driving, etc.

[0033] By adopting the scheme described in the above method embodiment, different methods can be used to determine the direction of the obstacle for the starting trajectory point and the non-starting trajectory point. For the starting trajectory point, the direction of the obstacle can be determined based on the predicted direction and the speed direction, and for each non-starting trajectory point, the direction of the obstacle can be determined in combination with the speed direction and the correction coefficient, thereby improving the accuracy of the determined direction, and then improving the accuracy of the subsequent processing results, and accordingly improving the safety of vehicle driving, etc.

[0034] For any obstacle detected, the predicted movement trajectory of the obstacle within a predetermined time period in the future can be obtained. The specific value of the predetermined time period can be determined according to actual needs, such as 6 seconds.

[0035] The moving trajectory may include a series of continuous trajectory points, wherein the first trajectory point may be called the starting trajectory point, that is, the current location of the obstacle, and the other trajectory points are non-starting trajectory points.

[0036] For the starting trajectory point, the predicted direction and speed direction of the obstacle at the starting trajectory point can be obtained, and the target direction of the obstacle at the starting trajectory point can be determined based on the obtained information.

[0037] Preferably, the vehicle may be an autonomous driving vehicle, and accordingly, the orientation information of the obstacle at the starting trajectory point generated by a perception module in the autonomous driving vehicle may be obtained as the predicted orientation.

[0038] Traditionally, the perception module in an autonomous driving vehicle will give the current direction and speed of each obstacle, that is, the direction information of each obstacle at its current location (starting trajectory point) and the speed direction of each obstacle at its current location. The direction information can be used as a predicted direction. In addition, for ease of expression, the speed direction can be referred to as a first reference direction.

[0039] With the help of the perception module, the required predicted direction and speed direction can be easily and conveniently obtained, thus laying a good foundation for subsequent processing.

[0040] The predicted direction and the first reference direction may be inconsistent. For example, taking the obstacle as a vehicle, the reasons for the inconsistency may include: 1. The perception module cannot distinguish between the front and rear of the vehicle, so the direction information of the obstacle given is reversed; 2. The vehicle is in the process of turning. Since the vehicle body has a certain length, the vehicle is controlled by the front wheels. Therefore, the change in the direction of the vehicle (i.e., the change in the direction of the vehicle body) lags behind the change in the speed direction; 3. Perception error. Figure 2 This is a schematic diagram of the relationship between the vehicle body orientation and the speed direction during the vehicle turning process described in the present disclosure. It can be seen that the change in the vehicle body orientation usually lags behind the change in the speed direction.

[0041] Preferably, a first deviation angle between the predicted orientation and the first reference direction can be obtained. In response to the first deviation angle being less than or equal to a first threshold, the target orientation of the obstacle at the starting trajectory point can be determined based on the first deviation angle. In response to the first deviation angle being greater than the first threshold, the predicted orientation can be corrected in a predetermined manner, and a second deviation angle between the corrected predicted orientation and the first reference direction can be obtained, and the target orientation of the obstacle at the starting trajectory point can be determined based on the second deviation angle.

[0042] In addition, preferably, the first threshold value may be 90 degrees, and the method of correcting the predicted orientation in a predetermined manner may include: adding a correction amount of 180 degrees to the predicted orientation. If the first deviation angle is greater than 90 degrees, it usually means that the vehicle orientation is opposite to the speed direction, and accordingly, a correction amount of 180 degrees may be added to the predicted orientation, thereby restoring the vehicle orientation to the positive direction.

[0043] That is to say, if the deviation angle between the predicted direction and the first reference direction (referred to as the first deviation angle for ease of distinction) is less than or equal to 90 degrees, the target direction of the obstacle at the starting trajectory point can be determined directly based on the first deviation angle; if the first deviation angle is greater than 90 degrees, a correction of 180 degrees can be added to the predicted direction and then the deviation angle can be calculated to obtain the second deviation angle, and then the target direction of the obstacle at the starting trajectory point can be determined based on the second deviation angle.

[0044] It can be seen that in the above processing method, the first deviation angle can be compared with the first threshold, and whether the predicted direction needs to be corrected can be determined based on the comparison result, thereby improving the accuracy of the predicted direction, and further improving the accuracy of the subsequently determined target direction.

[0045] Preferably, the method for determining the target orientation of the obstacle at the starting trajectory point based on the first deviation angle may include: in response to the first deviation angle being greater than the second threshold, using the first reference direction as the target orientation of the obstacle at the starting trajectory point, and in response to the first deviation angle being less than or equal to the second threshold, using the predicted orientation as the target orientation of the obstacle at the starting trajectory point, the second threshold being less than the first threshold; similarly, the method for determining the target orientation of the obstacle at the starting trajectory point based on the second deviation angle may include: in response to the second deviation angle being greater than the second threshold, using the first reference direction as the target orientation of the obstacle at the starting trajectory point, and in response to the second deviation angle being less than or equal to the second threshold, using the corrected predicted orientation as the target orientation of the obstacle at the starting trajectory point.

[0046] The specific value of the second threshold can be determined according to actual needs, for example, it can be 36 degrees.

[0047] Whether it is the first deviation angle or the second deviation angle, if it is less than or equal to 36 degrees, the predicted direction or the corrected predicted direction can be considered to be credible, so that the predicted direction or the corrected predicted direction can be used as the target direction of the obstacle at the starting trajectory point. Conversely, if it is greater than 36 degrees, the predicted direction can be considered to be unreliable, and accordingly, the first reference direction can be used as the target direction of the obstacle at the starting trajectory point.

[0048] In the above processing method, whether the predicted direction is credible can be judged, and then the target direction of the obstacle at the starting trajectory point can be determined in corresponding ways according to different judgment results, thereby improving the accuracy of the determined target direction.

[0049] In addition to determining the target orientation at the starting trajectory point, a correction coefficient for the orientation can also be determined.

[0050] Preferably, the obstacle includes: a vehicle; the method of determining the correction coefficient may include: in response to taking the first reference direction as the target orientation of the obstacle at the starting trajectory point, setting the correction coefficient to 1, in response to taking the predicted orientation or the corrected predicted orientation as the target orientation of the obstacle at the starting trajectory point, obtaining the turning radius of curvature at the starting trajectory point, determining a first estimated deviation angle according to the turning radius of curvature, the vehicle length of the obstacle and the kinematic model, the first estimated deviation angle being the deviation angle between the estimated target orientation of the obstacle at the starting trajectory point and the first reference direction, and determining the correction coefficient according to the first estimated deviation angle.

[0051] Preferably, the method of determining the correction coefficient based on the first estimated deviation angle may include: in response to the predicted direction not being corrected, obtaining the ratio of the first deviation angle to the first estimated deviation angle; in response to the predicted direction being corrected, obtaining the ratio of the second deviation angle to the first estimated deviation angle, and using the ratio of the first deviation angle to the first estimated deviation angle or the ratio of the second deviation angle to the first estimated deviation angle as a reference ratio; in response to the reference ratio being a negative number, setting the correction coefficient to 0; in response to the reference ratio being a positive number, performing the following processing: in response to the reference ratio being less than a third threshold, setting the correction coefficient to the third threshold; in response to the reference ratio being greater than a fourth threshold, setting the correction coefficient to the fourth threshold; in response to the reference ratio being greater than or equal to the third threshold and less than or equal to the fourth threshold, setting the correction coefficient to the reference ratio, the fourth threshold being greater than the third threshold.

[0052] The specific values ​​of the third threshold and the fourth threshold can be determined according to actual conditions. For example, the third threshold can be 0.3, and the fourth threshold can be 2.0.

[0053] If the first reference direction is used as the target orientation of the obstacle at the starting trajectory point, that is, if the predicted orientation is unreliable, then the correction coefficient can be set to 1, that is, the initial correction amount described later is fully trusted. Otherwise, the first estimated deviation angle can be determined, and the ratio of the first deviation angle or the second deviation angle to the first estimated deviation angle can be obtained, wherein, if the predicted orientation is not corrected, the ratio of the first deviation angle to the first estimated deviation angle is obtained, and if the predicted orientation is corrected, the ratio of the second deviation angle to the first estimated deviation angle is obtained, that is, the reference ratio = the first deviation angle Or the second deviation angle / the first estimated deviation angle, if the reference ratio is a negative number, it means that the estimated vehicle body orientation is opposite to the actual vehicle body orientation and is wrong, so there is no need to use the subsequent initial correction amount. Accordingly, the correction coefficient can be set to 0. If the reference ratio is a positive number and is less than 0.3, the correction coefficient can be set to 0.3. If it is greater than 2.0, the correction coefficient can be set to 2.0. Otherwise, the correction coefficient can be directly set to the reference ratio, that is, the correction coefficient is limited to between 0.3 and 2.0, thereby improving the rationality of the correction.

[0054] As mentioned above, when the obstacle is a vehicle, the turning radius of curvature at the starting trajectory point can be obtained, and the first estimated deviation angle can be determined according to the turning radius of curvature, the vehicle length of the obstacle and the kinematic model. The vehicle length can be sensed.

[0055] Preferably, the ratio of the vehicle length to the steering curvature radius can be obtained, and the product of the ratio and a predetermined coefficient can be obtained, and then the arc tangent (arctan) operation result of the product can be obtained, and the arc tangent operation result is used as the required first estimated deviation angle. The specific value of the predetermined coefficient can be determined according to actual needs, such as 0.5.

[0056] Accordingly, we have:

[0057] First estimated deviation angle = arctan (0.5*vehicle length / steering curvature radius); (1)

[0058] The first estimated deviation angle may not be completely accurate. The reasons may include: 1. The perceived vehicle length is not accurate enough, especially for a large vehicle that is partially obscured; 2. The obtained steering curvature radius may have errors.

[0059] Take the obstacle as a vehicle as an example. Figure 3 Schematic diagram of the relationship between the angle between the vehicle body direction and the speed direction and the turning curvature radius during the vehicle turning process described in the present disclosure. Figure 3 As shown, L represents the vehicle length, R represents the steering curvature radius, and θ represents the first estimated deviation angle.

[0060] In the above processing method, with the help of the kinematic model, the required first estimated deviation angle can be determined efficiently and quickly, thus laying a good foundation for subsequent processing.

[0061] Preferably, the method for acquiring the turning curvature radius may include: determining the turning curvature at the starting trajectory point by a three-point method, and determining the turning curvature radius according to the turning curvature.

[0062] For example, the turning curvature at the starting trajectory point can be determined by the three-point method based on the starting trajectory point and the two trajectory points after the starting trajectory point, and then the inverse of the turning curvature can be used as the turning curvature radius. The three-point method is a mature existing method, which is simple to implement and has good accuracy.

[0063] After determining the target direction and correction coefficient of the obstacle at the starting trajectory point, the target direction at each non-starting trajectory point can also be determined, so that the speed direction of the obstacle at the non-starting trajectory point can be obtained as the second reference direction, and the initial correction amount corresponding to the non-starting trajectory point can be determined. The target direction of the obstacle at the non-starting trajectory point is determined based on the initial correction amount, the correction coefficient and the second reference direction. For the sake of distinction, the speed direction of the obstacle at the non-starting trajectory point is called the second reference direction.

[0064] Preferably, the obstacle is a vehicle. Accordingly, the turning radius of curvature at a non-starting trajectory point can be obtained. According to the turning radius of curvature, the vehicle length of the obstacle and the kinematic model, a second estimated deviation angle is determined as an initial correction amount. The second estimated deviation angle is the deviation angle between the estimated target orientation of the obstacle at the non-starting trajectory point and the second reference direction. Then, the target orientation of the obstacle at the non-starting trajectory point can be determined according to the initial correction amount, the correction coefficient and the second reference direction.

[0065] Preferably, a method for determining the target orientation of an obstacle at a non-starting trajectory point based on an initial correction amount, a correction coefficient, and a second reference direction may include: obtaining the product of the initial correction amount and the correction coefficient as the target correction amount, adding the second reference direction to the target correction amount, and obtaining the target orientation of the obstacle at the non-starting trajectory point.

[0066] That is:

[0067] Target correction amount = initial correction amount * correction coefficient;

[0068] Target orientation = second reference direction + target correction; (2)

[0069] The target orientation refers to the target orientation of the obstacle at a non-starting trajectory point. In addition, the correction coefficient may be 1, 0, or any value between 0.3 and 2.0.

[0070] In the above processing method, the initial correction amount can be determined according to the turning curvature radius, the vehicle length of the obstacle and the kinematic model, and the initial correction amount can be further corrected using the correction coefficient to obtain the target correction amount, and then the final required target orientation can be determined in combination with the speed direction of the obstacle at the non-starting trajectory point and the target correction amount, thereby improving the accuracy of the determined target orientation.

[0071] When the obstacle is a vehicle, preferably, for any non-starting trajectory point, the ratio of the vehicle length to the steering curvature radius at the non-starting trajectory point can be obtained, and the product of the ratio and 0.5 can be obtained, and then the inverse tangent operation result of the product can be obtained, and the inverse tangent operation result is used as the required second estimated deviation angle, which can be specifically referred to as shown in formula (1).

[0072] In addition, preferably, for any non-starting trajectory point, the method for obtaining the turning curvature radius may include: determining the turning curvature at the non-starting trajectory point by a three-point method, and determining the turning curvature radius according to the turning curvature.

[0073] For example, for any non-starting track point, the turning curvature at the non-starting track point can be determined by the three-point method based on the non-starting track point and the two track points before and after the non-starting track point, and then the inverse of the turning curvature can be used as the turning curvature radius. In particular, if the non-starting track point is the last non-starting track point, the turning curvature at the non-starting track point can be determined by the three-point method in combination with the non-starting track point and the two track points before it.

[0074] In practical applications, after the turning curvature at each trajectory point (including the starting trajectory point and each non-starting trajectory point) is determined by the three-point method, each turning curvature can be used to form a curvature sequence in time sequence, and then each turning curvature in the curvature sequence can be smoothed and filtered, and then each turned curvature after processing can be used as the turning curvature at each trajectory point, and the turning curvature at each trajectory point can be optimized to improve the quality of each turning curvature and avoid mutations. The specific method of the smoothing filter can be mean filtering, etc.

[0075] Preferably, for any non-starting trajectory point, after determining the target orientation of the obstacle at the non-starting trajectory point based on the second reference direction and the correction coefficient, the non-starting trajectory point can also be used as a trajectory point to be processed, and the target orientation at the previous trajectory point adjacent to the trajectory point to be processed can be used as a reference orientation, and the target orientation of the obstacle at the trajectory point to be processed can be corrected according to the reference orientation.

[0076] Preferably, in response to the target orientation of the obstacle at the trajectory point to be processed being less than the difference between the reference orientation and the fifth threshold, the difference can be used as the target orientation of the obstacle at the trajectory point to be processed; in response to the target orientation of the obstacle at the trajectory point to be processed being greater than the sum of the reference orientation and the fifth threshold, the sum can be used as the target orientation of the obstacle at the trajectory point to be processed; in response to the target orientation of the obstacle at the trajectory point to be processed being greater than or equal to the difference and less than or equal to the sum, the target orientation of the obstacle at the trajectory point to be processed can be maintained unchanged.

[0077] The specific value of the fifth threshold can be determined according to actual needs.

[0078] Through the above processing, the target orientation of the obstacle at the trajectory point to be processed can be limited to [reference orientation-fifth threshold, reference orientation+fifth threshold], thereby limiting the change rate of the target orientation and avoiding sudden changes in the target orientation as much as possible, thereby reducing problems such as obstacle projection errors in the planning stage caused by sudden changes in orientation.

[0079] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are expressed as a series of action combinations, but those skilled in the art should be aware that the present disclosure is not limited by the described order of actions, because according to the present disclosure, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present disclosure.

[0080] The above is an introduction to the method embodiment. The following is a further explanation of the scheme disclosed in the present invention through an apparatus embodiment.

[0081] Figure 4 FIG. 4 is a schematic diagram of the structure of the obstacle direction acquisition device embodiment 400 described in the present disclosure. Figure 4 As shown, it includes: a trajectory acquisition module 401 , a first determination module 402 and a second determination module 403 .

[0082] The trajectory acquisition module 401 is used to obtain the predicted movement trajectory of any obstacle detected during the vehicle's driving within a predetermined time period in the future. The movement trajectory consists of a starting trajectory point and non-starting trajectory points. The starting trajectory point is the current location of the obstacle.

[0083] The first determination module 402 is used to obtain the speed direction of the obstacle at the starting trajectory point as the first reference direction, and obtain the predicted orientation of the obstacle at the starting trajectory point, and determine the target orientation of the obstacle at the starting trajectory point according to a first deviation angle between the predicted orientation and the first reference direction, wherein the target orientation is the orientation determined as the desired result.

[0084] The second determination module 403 is used to obtain the speed direction of the obstacle at each non-starting trajectory point as the second reference direction, and determine the initial correction amount corresponding to the non-starting trajectory point, and determine the target direction of the obstacle at the non-starting trajectory point according to the initial correction amount, the correction coefficient and the second reference direction, wherein the correction coefficient is determined according to the target direction at the starting trajectory point.

[0085] By adopting the scheme described in the above-mentioned device embodiment, different methods can be used to determine the direction of the obstacle for the starting trajectory point and the non-starting trajectory point. For the starting trajectory point, the direction of the obstacle can be determined based on the predicted direction and the speed direction, etc. For each non-starting trajectory point, the direction of the obstacle can be determined in combination with the speed direction and the correction coefficient, etc., thereby improving the accuracy of the determined direction, and then improving the accuracy of the subsequent processing results, and accordingly improving the safety of vehicle driving, etc.

[0086] Preferably, the vehicle may be an autonomous driving vehicle, and accordingly, the first determination module 402 may obtain the orientation information of the obstacle at the starting trajectory point generated by the perception module in the autonomous driving vehicle as the predicted orientation.

[0087] Preferably, after obtaining the first deviation angle between the predicted orientation and the first reference direction, the first determination module 402 determines the target orientation of the obstacle at the starting trajectory point based on the first deviation angle in response to the first deviation angle being less than or equal to the first threshold, and corrects the predicted orientation in a predetermined manner in response to the first deviation angle being greater than the first threshold, and obtains the second deviation angle between the corrected predicted orientation and the first reference direction, and determines the target orientation of the obstacle at the starting trajectory point based on the second deviation angle.

[0088] In addition, preferably, the first threshold may be 90 degrees, and the predetermined method may include: adding a correction amount of 180 degrees to the predicted direction.

[0089] Preferably, the way in which the first determination module 402 determines the target orientation of the obstacle at the starting trajectory point according to the first deviation angle may include: in response to the first deviation angle being greater than the second threshold, using the first reference direction as the target orientation of the obstacle at the starting trajectory point; in response to the first deviation angle being less than or equal to the second threshold, using the predicted orientation as the target orientation of the obstacle at the starting trajectory point, the second threshold being less than the first threshold; similarly, the way in which the target orientation of the obstacle at the starting trajectory point according to the second deviation angle may include: in response to the second deviation angle being greater than the second threshold, using the first reference direction as the target orientation of the obstacle at the starting trajectory point; in response to the second deviation angle being less than or equal to the second threshold, using the corrected predicted orientation as the target orientation of the obstacle at the starting trajectory point.

[0090] Preferably, the second threshold value may be 36 degrees. Regardless of whether it is the first deviation angle or the second deviation angle, if it is less than or equal to 36 degrees, the predicted orientation or the corrected predicted orientation may be considered to be credible, so that the predicted orientation or the corrected predicted orientation may be used as the target orientation of the obstacle at the starting trajectory point. Conversely, if it is greater than 36 degrees, the predicted orientation may be considered to be unreliable, and accordingly, the first reference direction may be used as the target orientation of the obstacle at the starting trajectory point.

[0091] In addition to determining the target orientation at the starting trajectory point, a correction coefficient for the orientation can also be determined.

[0092] Preferably, the obstacle includes: a vehicle; the way in which the first determination module 402 determines the correction coefficient may include: in response to taking the first reference direction as the target orientation of the obstacle at the starting trajectory point, setting the correction coefficient to 1, in response to taking the predicted orientation or the corrected predicted orientation as the target orientation of the obstacle at the starting trajectory point, obtaining the turning radius of curvature at the starting trajectory point, determining a first estimated deviation angle according to the turning radius of curvature, the vehicle length of the obstacle and the kinematic model, the first estimated deviation angle being the deviation angle between the estimated target orientation of the obstacle at the starting trajectory point and the first reference direction, and determining the correction coefficient according to the first estimated deviation angle.

[0093] Preferably, the first determination module 402 determines the correction coefficient according to the first estimated deviation angle in a manner that may include: in response to not correcting the predicted direction, obtaining the ratio of the first deviation angle to the first estimated deviation angle; in response to correcting the predicted direction, obtaining the ratio of the second deviation angle to the first estimated deviation angle, and using the ratio of the first deviation angle to the first estimated deviation angle or the ratio of the second deviation angle to the first estimated deviation angle as a reference ratio; in response to the reference ratio being a negative number, setting the correction coefficient to 0; in response to the reference ratio being a positive number, performing the following processing: in response to the reference ratio being less than a third threshold, setting the correction coefficient to the third threshold; in response to the reference ratio being greater than a fourth threshold, setting the correction coefficient to the fourth threshold; in response to the reference ratio being greater than or equal to the third threshold and less than or equal to the fourth threshold, setting the correction coefficient to the reference ratio, and the fourth threshold being greater than the third threshold.

[0094] As mentioned above, when the obstacle is a vehicle, the turning radius of curvature at the starting trajectory point can be obtained, and the first estimated deviation angle can be determined according to the turning radius of curvature, the vehicle length of the obstacle and the kinematic model. The vehicle length can be sensed.

[0095] Preferably, the first determination module 402 can obtain the ratio of the vehicle length to the steering curvature radius, and can obtain the product of the ratio and a predetermined coefficient, and further obtain the inverse tangent operation result of the product, and use the inverse tangent operation result as the required first estimated deviation angle.

[0096] In addition, the first determination module 402 may determine the turning curvature at the starting trajectory point by a three-point method, and determine the turning curvature radius according to the turning curvature.

[0097] After determining the target orientation and correction coefficient of the obstacle at the starting trajectory point, the target orientation at each non-starting trajectory point can also be determined respectively.

[0098] Preferably, the obstacle may be a vehicle. Accordingly, the second determination module 403 may use the speed direction of the obstacle at the non-starting trajectory point as the second reference direction, and may obtain the turning radius of curvature at the non-starting trajectory point, and determine the second estimated deviation angle as an initial correction amount according to the turning radius of curvature, the vehicle length of the obstacle and the kinematic model. The second estimated deviation angle is the deviation angle between the estimated target orientation of the obstacle at the non-starting trajectory point and the second reference direction, and then the target orientation of the obstacle at the non-starting trajectory point may be determined according to the initial correction amount, the correction coefficient and the second reference direction.

[0099] Preferably, the second determination module 403 determines the target orientation of the obstacle at the non-starting trajectory point based on the initial correction amount, the correction coefficient and the second reference direction, which may include: obtaining the product of the initial correction amount and the correction coefficient as the target correction amount, adding the second reference direction and the target correction amount, and obtaining the target orientation of the obstacle at the non-starting trajectory point.

[0100] When the obstacle is a vehicle, preferably, the second determination module 403 can obtain, for any non-starting trajectory point, the ratio of the vehicle length to the steering curvature radius at the non-starting trajectory point, and can obtain the product of the ratio and a predetermined coefficient, and then can obtain the inverse tangent operation result of the product, and use the inverse tangent operation result as the required second estimated deviation angle.

[0101] In addition, preferably, for any non-starting trajectory point, the second determining module 403 can determine the turning curvature at the non-starting trajectory point by a three-point method, and determine the turning curvature radius according to the turning curvature.

[0102] Preferably, the second determination module 403, for any non-starting trajectory point, after determining the target orientation of the obstacle at the non-starting trajectory point, can also use the non-starting trajectory point as a trajectory point to be processed, and can use the target orientation at the previous trajectory point adjacent to the trajectory point to be processed as a reference orientation, and correct the target orientation of the obstacle at the trajectory point to be processed according to the reference orientation.

[0103] Preferably, in response to the target orientation of the obstacle at the trajectory point to be processed being less than the difference between the reference orientation and the fifth threshold, the second determination module 403 may use the difference as the target orientation of the obstacle at the trajectory point to be processed; in response to the target orientation of the obstacle at the trajectory point to be processed being greater than the sum of the reference orientation and the fifth threshold, the sum may be used as the target orientation of the obstacle at the trajectory point to be processed; in response to the target orientation of the obstacle at the trajectory point to be processed being greater than or equal to the difference and less than or equal to the sum, the target orientation of the obstacle at the trajectory point to be processed may be maintained unchanged.

[0104] Figure 4 The specific working process of the illustrated device embodiment can refer to the relevant description in the aforementioned method embodiment and will not be described in detail.

[0105] In summary, the solution disclosed in the present invention can improve the accuracy of the direction of the determined obstacle, thereby improving driving safety, etc.

[0106] The solution disclosed in the present disclosure can be applied to the field of artificial intelligence, especially to the fields of autonomous driving, computer vision and deep learning. Artificial intelligence is a discipline that studies how computers can simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.). It has both hardware-level and software-level technologies. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, and big data processing. Artificial intelligence software technologies mainly include computer vision technology, speech recognition technology, natural language processing technology, machine learning / deep learning, big data processing technology, knowledge graph technology, and other major directions.

[0107] The movement tracks and the like in the embodiments described in this disclosure are not for a specific user and do not reflect the personal information of a specific user. In the technical solution of this disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0108] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0109] Figure 5 A schematic block diagram of an electronic device 500 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0110] like Figure 5 As shown, the device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0111] A number of components in the device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0112] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as the methods described in the present disclosure. For example, in some embodiments, the methods described in the present disclosure may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of the method described in the present disclosure may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the method described in the present disclosure in any other appropriate manner (e.g., by means of firmware).

[0113] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0114] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0115] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, 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 foregoing.

[0116] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0117] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0118] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0119] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0120] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for obtaining an obstacle direction, comprising: For any obstacle detected during vehicle driving, obtain a predicted moving trajectory of the obstacle within a predetermined time period in the future, wherein the moving trajectory consists of a starting trajectory point and non-starting trajectory points, and the starting trajectory point is the current location of the obstacle; Acquire the speed direction of the obstacle at the starting trajectory point as a first reference direction, and acquire the predicted orientation of the obstacle at the starting trajectory point, and determine the target orientation of the obstacle at the starting trajectory point according to a first deviation angle between the predicted orientation and the first reference direction, wherein the target orientation is the orientation determined as a desired result; For each non-starting trajectory point, the speed direction of the obstacle at the non-starting trajectory point is obtained as the second reference direction, and an initial correction amount corresponding to the non-starting trajectory point is determined; the target direction of the obstacle at the non-starting trajectory point is determined according to the initial correction amount, the correction coefficient and the second reference direction, wherein the correction coefficient is determined according to the target direction at the starting trajectory point.

2. The method according to claim 1, wherein: The vehicle includes: an autonomous driving vehicle; The obtaining of the predicted orientation of the obstacle at the starting trajectory point includes: obtaining orientation information of the obstacle at the starting trajectory point generated by a perception module in the autonomous driving vehicle as the predicted orientation.

3. The method according to claim 1, wherein: Determining the target orientation of the obstacle at the starting trajectory point includes: In response to the first deviation angle being less than or equal to a first threshold, determining a target orientation of the obstacle at the starting trajectory point according to the first deviation angle; In response to the first deviation angle being greater than the first threshold, the predicted orientation is corrected in a predetermined manner, and a second deviation angle between the corrected predicted orientation and the first reference direction is obtained, and the target orientation of the obstacle at the starting trajectory point is determined based on the second deviation angle.

4. The method according to claim 3, wherein: The first threshold is 90 degrees; The predetermined method includes: adding a correction amount of 180 degrees to the predicted orientation.

5. The method according to claim 3, wherein: Determining the target orientation of the obstacle at the starting trajectory point according to the first deviation angle includes: in response to the first deviation angle being greater than a second threshold, using the first reference direction as the target orientation of the obstacle at the starting trajectory point; in response to the first deviation angle being less than or equal to the second threshold, using the predicted orientation as the target orientation of the obstacle at the starting trajectory point; the second threshold is less than the first threshold; Determining the target orientation of the obstacle at the starting trajectory point based on the second deviation angle includes: in response to the second deviation angle being greater than the second threshold, using the first reference direction as the target orientation of the obstacle at the starting trajectory point; in response to the second deviation angle being less than or equal to the second threshold, using the corrected predicted orientation as the target orientation of the obstacle at the starting trajectory point.

6. The method according to claim 5, wherein: The obstacles include: vehicles; The correction coefficient is determined by: In response to taking the first reference direction as the target orientation of the obstacle at the starting trajectory point, setting the correction coefficient to 1; In response to taking the predicted orientation or the corrected predicted orientation as the target orientation of the obstacle at the starting trajectory point, the turning radius of curvature at the starting trajectory point is obtained, and a first estimated deviation angle is determined based on the turning radius of curvature, the vehicle length of the obstacle, and a kinematic model. The first estimated deviation angle is the estimated deviation angle between the target orientation of the obstacle at the starting trajectory point and the first reference direction, and the correction coefficient is determined based on the first estimated deviation angle.

7. The method according to claim 6, wherein: Determining the first estimated deviation angle includes: obtaining the ratio of the vehicle length to the steering curvature radius, obtaining the product of the ratio and a predetermined coefficient, obtaining an inverse tangent operation result of the product, and using the inverse tangent operation result as the first estimated deviation angle.

8. The method according to claim 6, wherein: Determining the correction coefficient according to the first estimated deviation angle includes: In response to not correcting the predicted orientation, obtaining a ratio of the first deviation angle to the first estimated deviation angle; in response to correcting the predicted orientation, obtaining a ratio of the second deviation angle to the first estimated deviation angle, and using the ratio of the first deviation angle to the first estimated deviation angle or the ratio of the second deviation angle to the first estimated deviation angle as a reference ratio; In response to the reference ratio being a negative number, setting the correction coefficient to 0; In response to the reference ratio being a positive number, the following processing is performed: if the reference ratio is less than a third threshold, the correction coefficient is set to the third threshold; if the reference ratio is greater than a fourth threshold, the correction coefficient is set to the fourth threshold; if the reference ratio is greater than or equal to the third threshold and less than or equal to the fourth threshold, the correction coefficient is set to the reference ratio, and the fourth threshold is greater than the third threshold.

9. The method according to claim 1, wherein: The obstacles include: vehicles; The determining of the initial correction amount corresponding to the non-starting trajectory point comprises: The turning radius of curvature at the non-starting trajectory point is obtained, and a second estimated deviation angle is determined as the initial correction amount according to the turning radius of curvature, the vehicle length of the obstacle, and a kinematic model. The second estimated deviation angle is an estimated deviation angle between a target orientation of the obstacle at the non-starting trajectory point and the second reference direction.

10. The method according to claim 9, wherein: The determining the target orientation of the obstacle at the non-starting trajectory point according to the initial correction amount, the correction coefficient and the second reference direction comprises: Obtaining the product of the initial correction amount and the correction coefficient as a target correction amount; The second reference direction is added to the target correction amount to obtain the target orientation of the obstacle at the non-starting trajectory point.

11. The method according to claim 9, wherein: Determining the second estimated deviation angle includes: obtaining the ratio of the vehicle length to the steering curvature radius, obtaining the product of the ratio and a predetermined coefficient, obtaining an inverse tangent operation result of the product, and using the inverse tangent operation result as the second estimated deviation angle.

12. The method according to any one of claims 1 to 11, further comprising: After determining the target orientation of the obstacle at the non-starting trajectory point, the non-starting trajectory point is used as a trajectory point to be processed, and the target orientation at a previous trajectory point adjacent to the trajectory point to be processed is used as a reference orientation, and the target orientation of the obstacle at the trajectory point to be processed is corrected according to the reference orientation.

13. The method according to claim 12, wherein: The correcting the target orientation of the obstacle at the to-be-processed trajectory point according to the reference orientation comprises: In response to the target orientation of the obstacle at the to-be-processed trajectory point being less than the difference between the reference orientation and a fifth threshold, taking the difference as the target orientation of the obstacle at the to-be-processed trajectory point; In response to the target orientation of the obstacle at the to-be-processed trajectory point being greater than the sum of the reference orientation and a fifth threshold, taking the sum as the target orientation of the obstacle at the to-be-processed trajectory point; In response to the target orientation of the obstacle at the to-be-processed trajectory point being greater than or equal to the difference and less than or equal to the sum, the target orientation of the obstacle at the to-be-processed trajectory point is maintained unchanged.

14. An obstacle direction acquisition device, comprising: A trajectory acquisition module, a first determination module and a second determination module; The trajectory acquisition module is used to acquire the predicted moving trajectory of any obstacle detected during the driving process of the vehicle within a predetermined time period in the future, wherein the moving trajectory consists of a starting trajectory point and non-starting trajectory points, and the starting trajectory point is the current location of the obstacle; The first determination module is used to obtain the speed direction of the obstacle at the starting trajectory point as a first reference direction, and obtain the predicted orientation of the obstacle at the starting trajectory point, and determine the target orientation of the obstacle at the starting trajectory point according to a first deviation angle between the predicted orientation and the first reference direction, wherein the target orientation is the orientation determined as the desired result; The second determination module is used to obtain the speed direction of the obstacle at each non-starting trajectory point as the second reference direction, and determine the initial correction amount corresponding to the non-starting trajectory point, and determine the target direction of the obstacle at the non-starting trajectory point according to the initial correction amount, the correction coefficient and the second reference direction, wherein the correction coefficient is determined according to the target direction at the starting trajectory point.

15. The device according to claim 14, wherein: The vehicle includes: an autonomous driving vehicle; The first determination module obtains the orientation information of the obstacle at the starting trajectory point generated by the perception module in the autonomous driving vehicle as the predicted orientation.

16. The device according to claim 14, wherein: In response to the first deviation angle being less than or equal to a first threshold, the first determination module determines the target orientation of the obstacle at the starting trajectory point based on the first deviation angle; in response to the first deviation angle being greater than the first threshold, the first determination module corrects the predicted orientation in a predetermined manner, obtains a second deviation angle between the corrected predicted orientation and the first reference direction, and determines the target orientation of the obstacle at the starting trajectory point based on the second deviation angle.

17. The device according to claim 16, wherein: The first threshold is 90 degrees; The predetermined method includes: adding a correction amount of 180 degrees to the predicted orientation.

18. The device according to claim 16, wherein: In response to the first deviation angle being greater than a second threshold, the first determination module uses the first reference direction as the target orientation of the obstacle at the starting trajectory point, and in response to the first deviation angle being less than or equal to the second threshold, uses the predicted orientation as the target orientation of the obstacle at the starting trajectory point; The second threshold is less than the first threshold; In response to the second deviation angle being greater than the second threshold, the first determination module uses the first reference direction as the target orientation of the obstacle at the starting trajectory point, and in response to the second deviation angle being less than or equal to the second threshold, uses the corrected predicted orientation as the target orientation of the obstacle at the starting trajectory point.

19. The device according to claim 18, wherein: The obstacles include: vehicles; The first determination module is further used to, in response to taking the first reference direction as the target orientation of the obstacle at the starting trajectory point, set the correction coefficient to 1, in response to taking the predicted orientation or the corrected predicted orientation as the target orientation of the obstacle at the starting trajectory point, obtain the turning radius of curvature at the starting trajectory point, determine a first estimated deviation angle according to the turning radius of curvature, the vehicle length of the obstacle and a kinematic model, the first estimated deviation angle being the estimated deviation angle between the target orientation of the obstacle at the starting trajectory point and the first reference direction, and determine the correction coefficient according to the first estimated deviation angle.

20. The device according to claim 19, wherein The first determination module obtains the ratio of the vehicle length to the steering curvature radius, obtains the product of the ratio and a predetermined coefficient, obtains an inverse tangent operation result of the product, and uses the inverse tangent operation result as the first estimated deviation angle.

21. The device according to claim 19, wherein In response to the predicted orientation not being corrected, the first determination module obtains the ratio of the first deviation angle to the first estimated deviation angle; in response to the predicted orientation being corrected, the module obtains the ratio of the second deviation angle to the first estimated deviation angle; and uses the ratio of the first deviation angle to the first estimated deviation angle or the ratio of the second deviation angle to the first estimated deviation angle as a reference ratio; in response to the reference ratio being a negative number, the correction coefficient is set to 0; in response to the reference ratio being a positive number, the following processing is performed: in response to the reference ratio being less than a third threshold, the correction coefficient is set to the third threshold; in response to the reference ratio being greater than a fourth threshold, the correction coefficient is set to the fourth threshold; in response to the reference ratio being greater than or equal to the third threshold and less than or equal to the fourth threshold, the correction coefficient is set to the reference ratio, and the fourth threshold is greater than the third threshold.

22. The device according to claim 14, wherein: The obstacles include: vehicles; The second determination module obtains the turning curvature radius at the non-starting trajectory point, and determines a second estimated deviation angle as an initial correction amount based on the turning curvature radius, the vehicle length of the obstacle and the kinematic model. The second estimated deviation angle is an estimated deviation angle between the target orientation of the obstacle at the non-starting trajectory point and the second reference direction.

23. The device according to claim 22, wherein: The second determination module obtains the product of the initial correction amount and the correction coefficient as a target correction amount, adds the second reference direction to the target correction amount, and obtains a target orientation of the obstacle at the non-starting trajectory point.

24. The device according to claim 22, wherein: The second determination module obtains the ratio of the vehicle length to the steering curvature radius, obtains the product of the ratio and a predetermined coefficient, obtains an inverse tangent operation result of the product, and uses the inverse tangent operation result as the second estimated deviation angle.

25. The device according to any one of claims 14 to 24, wherein: The second determination module is further used to, after determining the target orientation of the obstacle at the non-starting trajectory point, use the non-starting trajectory point as a trajectory point to be processed, and use the target orientation at a previous trajectory point adjacent to the trajectory point to be processed as a reference orientation, and correct the target orientation of the obstacle at the trajectory point to be processed according to the reference orientation.

26. The device according to claim 25, wherein In response to the target orientation of the obstacle at the trajectory point to be processed being less than the difference between the reference orientation and the fifth threshold, the second determination module uses the difference as the target orientation of the obstacle at the trajectory point to be processed; in response to the target orientation of the obstacle at the trajectory point to be processed being greater than the sum of the reference orientation and the fifth threshold, uses the sum as the target orientation of the obstacle at the trajectory point to be processed; and in response to the target orientation of the obstacle at the trajectory point to be processed being greater than or equal to the difference and less than or equal to the sum, maintains the target orientation of the obstacle at the trajectory point to be processed unchanged.

27. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 13.

28. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to make a computer execute the method according to any one of claims 1 to 13.

29. A computer program product, comprising a computer program / instruction, which implements the method according to any one of claims 1 to 13 when executed by a processor.

Images