Obstacle Recognition Method, Device, Electronic Device, and Storage Medium

Abstract

Embodiments of the present disclosure disclose an obstacle recognition method, apparatus, electronic device, and storage medium. The method includes: obtaining trailer information of a detected trailer within the sensing range of a target vehicle, where the target vehicle includes a target tractor and a target trailer towed by the target tractor; determining the trailer type of the detected trailer according to the trailer information and the associated information of the target tractor, and / or determining the trailer type of the detected trailer according to the trailer information and the associated information of the target trailer; and determining whether the detected trailer is an obstacle affecting the driving of the target vehicle according to the trailer type of the detected trailer. The technical solution of the present disclosure improves the recognition accuracy of obstacles.

Description

Technical Field

[0001] The present disclosure relates to the field of autonomous driving technologies, and particularly to an obstacle recognition method, apparatus, electronic device, and storage medium. Background Art

[0002] In autonomous driving technologies, collision detection is a basic functional safety requirement. An autonomous driving vehicle with a collision detection function can perceive obstacles with potential safety risks in advance and perform operations such as lane changing to avoid or decelerating to stop, or take remedial measures after a physical collision, such as no longer performing path planning and control, to reduce secondary injuries.

[0003] Currently, the mainstream collision detection in the industry mainly relies on perception solutions based on lidar, cameras, or multi-sensor fusion. Compared with traditional pressure sensors based on collision strips, on the one hand, it can provide a large-range perception and detection ability with a smaller number of hardware components. On the other hand, relying on the upgrade of computing power and the improvement of algorithms, the flexibility and robustness of perception have been greatly improved. Most importantly, such perception increases the possibility of collision warning and can avoid risks before potential personal life and property safety hazards occur.

[0004] However, regardless of which solution is adopted, the target of perception is obstacles. Therefore, accurately identifying obstacles is a very crucial step. Summary of the Invention

[0005] To solve the above technical problems or at least partially solve the above technical problems, embodiments of the present disclosure provide an obstacle recognition method, apparatus, electronic device, and storage medium, which improve the recognition accuracy of obstacles.

[0006] In a first aspect, embodiments of the present disclosure provide an obstacle recognition method, which includes:

[0007] Obtaining trailer information of a detected trailer within the perception range of a target vehicle, where the target vehicle includes a target tractor and a target trailer towed by the target tractor;

[0008] Determining the trailer type of the detected trailer according to the trailer information and the associated information of the target tractor, and / or determining the trailer type of the detected trailer according to the trailer information and the associated information of the target trailer;

[0009] Determining whether the detected trailer is an obstacle affecting the driving of the target vehicle according to the trailer type of the detected trailer.

[0010] Second aspect, an embodiment of the present disclosure further provides an obstacle recognition device, which includes: an acquisition module, configured to acquire the trailer information of a detected trailer within the sensing range of a target vehicle, where the target vehicle includes a target tractor and a target trailer towed by the target tractor;

[0011] A first determination module, configured to determine the trailer type of the detected trailer according to the trailer information and the association information of the target tractor, and / or determine the trailer type of the detected trailer according to the trailer information and the association information of the target trailer;

[0012] A second determination module, configured to determine whether the detected trailer is an obstacle affecting the driving of the target vehicle according to the trailer type of the detected trailer.

[0013] Third aspect, an embodiment of the present disclosure further provides an electronic device, where the electronic device includes: one or more processors; a storage device, configured to store one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the obstacle recognition method as described above.

[0014] Fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the obstacle recognition method as described above is implemented.

[0015] An obstacle recognition method provided by an embodiment of the present disclosure determines the trailer type of the detected trailer according to the trailer information of the detected trailer and the association information of the target tractor, and / or determines the trailer type of the detected trailer according to the trailer information and the association information of the target trailer; determines whether the detected trailer is an obstacle affecting the driving of the target vehicle according to the trailer type of the detected trailer, thereby improving the recognition accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Combined with the drawings and referring to the following specific embodiments, the above and other features, advantages and aspects of the embodiments of the present disclosure will become more obvious. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic, and the original elements and elements are not necessarily drawn to scale.

[0017] Figure 1 is a flowchart of an obstacle recognition method in an embodiment of the present disclosure;

[0018] Figure 2 is a schematic diagram of a rectangle corresponding to the target tractor in an embodiment of the present disclosure;

[0019] Figure 3Schematic diagram of the relationship for determining the segmentation coefficient in an embodiment of the present disclosure;

[0020] Figure 4 Flowchart of collision detection processing in an embodiment of the present disclosure;

[0021] Figure 5 Schematic diagram of the structure of an obstacle recognition device in an embodiment of the present disclosure;

[0022] Figure 6 Schematic diagram of the structure of an electronic device in an embodiment of the present disclosure. Detailed implementation manners

[0023] The embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not used to limit the protection scope of the present disclosure.

[0024] It should be noted that concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependent relationships.

[0025] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0026] With the exploration and implementation of more and more autonomous driving application scenarios, the definition boundary of the "ego vehicle" is constantly expanding. For example, in the logistics scenario, the trailer towed by a logistics tractor with autonomous driving capabilities (hereinafter referred to as the ego vehicle trailer) is an extension of the "ego vehicle". The perception system should filter the ego vehicle trailer from the detected obstacle trailers and only retain the other vehicle trailers, or retain both, but it is necessary to clarify that they belong to different categories. However, in the logistics scenario, the ego vehicle trailer and the other vehicle trailer are often similar in appearance, and the sizes, shapes, etc. of the trailers towed by the same logistics tractor are also different. It is difficult to distinguish between the two from the perspective of end-to-end object recognition. It is often based on certain prior knowledge (such as information about the number, size of the trailers, and a certain area behind the logistics tractor), and the ego vehicle trailer is predicted from the perspective of the kinematic model, and the trailers that meet the conditions are classified into the ego vehicle trailer. However, in actual application scenarios, the movement paths of the trailers are unpredictable, which leads to the situation that the detection based on the aforementioned method is prone to misidentifying the ego vehicle trailer as the other vehicle trailer (hereinafter referred to as the misdetection of the ego vehicle trailer).

[0027] In view of the above problems, embodiments of the present disclosure provide an obstacle recognition method, which improves the recognition accuracy, specifically, it can accurately distinguish the trailer of the host vehicle and the trailer of other vehicles. Figure 1 It is a flowchart of an obstacle recognition method in an embodiment of the present disclosure. This method can be executed by an obstacle recognition device, which can be implemented in a software and / or hardware manner, and the device can be configured in an electronic device. As Figure 1 shown, the method may specifically include the following steps:

[0028] S110. Obtain the trailer information of the detected trailer within the sensing range of the target vehicle, where the target vehicle includes a target tractor and a target trailer towed by the target tractor.

[0029] The sensing range of the target vehicle is determined by the detection ranges of other sensing sensors such as the on-vehicle lidar and / or on-vehicle camera of the target vehicle. The detected trailer includes the trailer of the host vehicle towed by the target tractor and the trailer of other vehicles towed by the tractors of other vehicles.

[0030] The trailer information includes some information required to determine whether there is a collision risk between the detected trailer and the target tractor, such as the real-time position, movement direction, movement speed, etc. of the detected trailer.

[0031] S120. Determine the trailer type of the detected trailer according to the trailer information and the associated information of the target tractor, and / or determine the trailer type of the detected trailer according to the trailer information and the associated information of the target trailer.

[0032] Among them, the associated information of the target tractor includes but is not limited to: the size information of the target tractor (such as the length and width of the target tractor) and the position information (such as the coordinates of the four vertices of the target tractor).

[0033] Exemplarily, the determining the trailer type of the detected trailer according to the trailer information and the associated information of the target tractor includes:

[0034] Dividing the target tractor into a first part including the front of the vehicle and a second part including the rear of the vehicle according to the associated information of the target tractor; determining the trailer type of the detected trailer based on the first part and / or the second part and the trailer information.

[0035] Further, the dividing the target tractor into a first part including the front of the vehicle and a second part including the rear of the vehicle according to the associated information of the target tractor includes;

[0036] Determine the corresponding rectangle according to the width and length of the target tractor, and determine the four vertex coordinates of the rectangle according to the position information in the associated information of the target tractor; determine the area range of the first part and the area range of the second part according to the segmentation coefficient and the four vertex coordinates of the rectangle; wherein, the segmentation coefficient is determined according to the width, length, center point of the target tractor and the possible angle range of the obstacle.

[0037] Exemplarily, refer to Figure 2 a schematic diagram of a rectangle corresponding to the target tractor as shown, and the four vertex coordinates of this rectangle are in the order of the upper left vertex - the upper right vertex - the lower left

[0038]

[0039] vertex - the lower right vertex as follows: (x tl , y tl ), (x tr , y tr ), (x bl , y bl ), (x br , y br ). Divide the target tractor into a first part including the front of the vehicle and a second part including the rear of the vehicle according to the following expression.

[0040]

[0041] Among them, Rect front represents the area range of the first part, and the upper right vertex coordinate, the lower right vertex coordinate, the lower left vertex coordinate and the upper left vertex coordinate of the first part are (x_front tr , y_front tr ), (x_front br , y_front br ), (x_front bl , y_front bl ), (x_front tl , y_front tl ).

[0042] Rect rear represents the area range of the second part, and the upper right vertex coordinate, the lower right vertex coordinate, the lower left vertex coordinate and the upper left vertex coordinate of the second part are (x_rear tr , y_rear tr ), (x_rear br , y_rear br ), (x_rear bl, y_rear bl ), (x_rear tl , y_rear tl ).

[0043] r represents the segmentation coefficient, which is determined according to the width, length, center point of the target tractor, and the angular range θ where obstacles may appear. Refer to the schematic diagram showing the relationship for determining the segmentation coefficient as shown in Figure 3 . According to Figure 3 , the following relational expression can be obtained: where θ is the angular range where obstacles may appear, which is an empirical value, r represents the segmentation coefficient, w represents the width of the target tractor, h represents the length of the target tractor, and o represents the center point of the target tractor. Based on this relational expression, the segmentation coefficient r can be calculated.

[0044] Furthermore, the trailer type of the detected trailer is determined based on the first part and / or the second part and the trailer information.

[0045] Based on the trailer information, it is determined whether the detected trailer has a collision risk with the first part and whether it has a collision risk with the second part; if the detected trailer has a collision risk with the first part, the detected trailer is determined to be a trailer of another vehicle; if the detected trailer has a collision risk with the second part, the detected trailer is determined to be a self - vehicle trailer; correspondingly, determining whether the detected trailer is an obstacle affecting the driving of the target vehicle according to the trailer type of the detected trailer includes: if the detected trailer is a trailer of another vehicle, the detected trailer is determined to be an obstacle affecting the driving of the target vehicle. That is, the trailer type includes a trailer of another vehicle and a self - vehicle trailer.

[0046] Generally speaking, the ideological principle for determining the trailer type of the detected trailer based on the trailer information and the associated information of the target tractor is as follows: 1. Suppose the self - vehicle trailer is misdetected. When it appears within a certain range in the driving direction of the target tractor, there will be a safety risk. Therefore, logically, it should be classified as a trailer of another vehicle and treated as an obstacle. 2. It is less likely for a trailer of another vehicle to appear within a certain range behind the target tractor. Therefore, a misdetected self - vehicle trailer that appears within this range should be classified as a self - vehicle trailer. Based on the above two points, in this stage, the method of re - segmenting the target tractor is adopted, resulting in two parts: the front part and the rear part. The front part is the main body for detection. If a trailer collides with this part, it is considered to be caused by a trailer of another vehicle. The rear part plays a "buffering" role. If a trailer collides with this part, it is considered to be caused by a self - vehicle trailer.

[0047] Further, determining the trailer type of the detected trailer according to the trailer information and the associated information of the target trailer includes:

[0048] Determining one or more of the morphological features, size features, and motion features of the detected trailer according to the trailer information; determining the trailer type of the detected trailer according to one or more of the morphological features, size features, and motion features, and the associated information of the target trailer.

[0049] Exemplarily, the morphological feature includes the area of the detected trailer, the size feature includes the maximum side length of the detected trailer, and the motion feature includes the magnitude and direction of the speed of the detected trailer; the associated information of the target trailer includes the area of the target trailer, the maximum side length of the target trailer, the magnitude of the speed of the target trailer, the direction of the speed of the target trailer, and the appearance area of the target trailer. Correspondingly, determining the trailer type of the detected trailer according to one or more of the morphological features, size features, and motion features, and the associated information of the target trailer includes:

[0050] Constructing influencing factors according to one or more of the morphological features, size features, and motion features, and the associated information of the target trailer; determining the trailer type of the detected trailer according to the influencing factors; wherein, the influencing factors include any one or more of the following: the first ratio of the area of the detected trailer to the area of the target trailer, the second ratio of the maximum side length of the detected trailer to the maximum side length of the target trailer, the third ratio of the magnitude of the speed of the detected trailer to the magnitude of the speed of the target trailer, the fourth ratio of the direction of the speed of the detected trailer to the direction of the speed of the target trailer, and the result information on whether the geometric center of the detected trailer is a point within the appearance area of the target trailer (this result information includes two cases, one is that the geometric center of the detected trailer is a point within the appearance area of the target trailer, and the other is that the geometric center of the detected trailer is not a point within the appearance area of the target trailer).

[0051] Further, determining the trailer type of the detected trailer according to the influencing factors includes:

[0052] Determine the first data for characterizing the trailer type of the detected trailer according to the first ratio; determine the second data for characterizing the trailer type of the detected trailer according to the second ratio; determine the third data for characterizing the trailer type of the detected trailer according to the third ratio; determine the fourth data for characterizing the trailer type of the detected trailer according to the fourth ratio; determine the fifth data for characterizing the trailer type of the detected trailer according to the result information on whether the geometric center of the detected trailer is a point within the appearance area of the target trailer; perform a weighted sum on the first data, the second data, the third data, the fourth data, and the fifth data according to the weights respectively matching the first ratio, the second ratio, the third ratio, the fourth ratio, and the result information to obtain a reference value; determine the trailer type of the detected trailer according to the parameter value and a preset threshold.

[0053] Specifically, the ideological principle for determining the trailer type of the detected trailer according to the trailer information and the associated information of the target trailer is as follows: 1. False detection of the self-vehicle trailer is usually a small part of the misdetected trailers. The resulting obstacles have significant morphological differences compared to the actual trailers of other vehicles. 2. The motion state of the self-vehicle trailer is highly correlated with the self-vehicle (i.e., the target tractor), while the motion state of the trailers of other vehicles has less correlation, or even negative correlation.

[0054] To define the morphological differences, two factors, namely the area size and the maximum side length, are used in the embodiments of the present invention. Among them, the area is a basic morphological feature. Since the self-vehicle trailer is rarely completely exposed within the detection range when misdetected, the size of the detected object generated by the misdetection of the self-vehicle trailer is much smaller than that of the trailers of other vehicles. Therefore, only by comparing the areas can the two be distinguished:

[0055]

[0056] where detected area and prior_knowledge area respectively refer to the area of the detected object (specifically the trailer obstacle here), that is, the area of the detected trailer, and the prior knowledge of the actual size of the trailer, that is, the area of the target trailer. f area is the first ratio.

[0057] However, the actual perception results are fluctuating and noisy, and the pose of the trailers of other vehicles is also uncertain. It is very likely that part of the body of the trailers of other vehicles enters the perception blind area. Therefore, there may be a situation where the area of the misdetected self-vehicle trailer is similar to that of the detected trailers of other vehicles. At this time, relying solely on the area size is not reliable. However, after an object is occluded, its one-dimensional information is usually retained. Therefore, the side length information is also taken into consideration:

[0058]

[0059] Among them, detected longest_side and prior_knowldege longest_side respectively refer to the maximum side length of the detected object (specifically the trailer obstacle here, i.e., the maximum side length of the detected trailer) and the prior knowledge of the actual maximum side length of the trailer (i.e., the second ratio of the maximum side length of the target trailer), f longest_side is the second ratio.

[0060] On the other hand, the motion state uses speed, direction, and trajectory information. The speed magnitude of the self-vehicle trailer is the same as that of the self-vehicle (i.e., the target tractor):

[0061]

[0062] Among them, detected velocity and vehicle velocity respectively refer to the speed magnitude of the detected object (specifically the trailer obstacle here, i.e., the speed magnitude of the detected trailer) and the speed magnitude of the vehicle itself (i.e., the speed magnitude of the target trailer, and the speed magnitude of the target tractor can be considered as the speed magnitude of the target trailer), f velocity is the third ratio.

[0063] However, due to the non-rigid connection, the speed direction may be quite different from that of the self-vehicle. But for the case where the speed direction of an obstacle trailer has a difference of more than 90° from the self-vehicle, it is very likely that it can also be classified as the trailer of another vehicle:

[0064]

[0065] Among them, detected direction and vehicle direction respectively refer to the speed direction of the detected object (specifically the trailer obstacle here) and the speed direction of the vehicle itself, f direction is the fourth ratio.

[0066] Next is the driving trajectory. Since it is difficult for an obstacle to be stably perceived in multiple consecutive frames, and it is easy to lose tracking and have a small number of detection frames, calculating the trajectory correlation between the obstacle and the vehicle is not very stable and reliable. However, through the kinematic model relationship between the target tractor and the self-vehicle trailer, the possible area where the self-vehicle trailer may appear at any time can be determined. Then, as long as it is judged whether the perceived obstacle trailer is within this area, the self-vehicle trailer and the trailer of another vehicle can be distinguished:

[0067] f trajectory= point_in_polygon((x, y), polygon)

[0068] Among them, (x, y) represents the geometric center of the detected object (specifically the trailer obstacle here), polygon represents the polygonal area where the predicted trailer of the host vehicle should appear, and point_in_polygon is a function to judge whether a point is inside a polygon, which can be composed of any computational geometry algorithm.

[0069] The factors utilized by the present invention are elaborated above. However, it is difficult for any single factor to determine whether the trailer of the obstacle is the trailer of another vehicle, or the recognition accuracy is not high because each factor is interfered by noise. Therefore, after all factors are collected, the final judgment result of the trailer type can be generated through a certain decision-making. Essentially, it is a binary classification problem. The present invention uses a scoring mechanism based on weights. First, a weight coefficient w x is assigned to each factor, where x is the factor subscript, representing the importance of the factor to the final result. If it is determined based on a certain factor that the trailer of the obstacle is the trailer of another vehicle, the score of this factor is the value of its weight coefficient; otherwise, no score is given. That is, if it is determined according to the first ratio that the trailer type of the detected trailer is the trailer of another vehicle, the first data used to characterize the trailer type of the detected trailer can be 1; otherwise, the first data is 0. The second data used to characterize the trailer type of the detected trailer is determined according to the second ratio; the third data used to characterize the trailer type of the detected trailer is determined according to the third ratio; the fourth data used to characterize the trailer type of the detected trailer is determined according to the fourth ratio; and the fifth data used to characterize the trailer type of the detected trailer is determined according to the result information of whether the geometric center of the detected trailer is a point within the appearance area of the target trailer, which is similar to the determination method of the above first data.

[0070] Thus, the scores generated by all factors are T. Only when T satisfies T > T obst will it be determined at this stage that the trailer of the obstacle is the trailer of another vehicle. The values of w and T obst can be determined by using machine learning methods, such as SVM, deep learning, etc., after obtaining multiple sets of test data.

[0071] S130. Determine whether the detected trailer is an obstacle affecting the driving of the target vehicle according to the trailer type of the detected trailer.

[0072] The obstacle recognition method provided in this embodiment is a method for identifying the trailer of other vehicles based on a post-processing method, which solves the problem of misdetection of the self-vehicle trailer caused by the similarity of trailers in the logistics scenario, thereby improving the accuracy of collision detection and reducing the problem of rough driving of autonomous driving caused thereby. Specifically, the present invention judges the trailer type in two stages. On the one hand, considering that the degree of damage caused by a collision to the target tractor and the target trailer is different, especially the target tractor requires stricter requirements, so a redundancy mechanism is provided; on the other hand, the definitions of obstacles for the two are different. Logically, any object that affects the normal driving of the target vehicle is an obstacle to the target vehicle, even if it may be a part of the whole to which the target vehicle belongs.

[0073] It can be understood that the embodiment of the present invention can not only solve the problems faced by autonomous driving collision detection in the logistics scenario, but also be extended to other scenarios troubled by object similarity. In addition, through the training and adjustment of parameters, it can also cope with the changes and differences that occur in the same scenario or different scenarios, with high flexibility.

[0074] The present invention can reduce the unexpected planning anomalies caused by misdetection of the self-vehicle trailer in collision detection, such as lane change avoidance or parking, etc., and reduce the rough driving of autonomous driving; the present invention has high expandability and flexibility. Vertically, parameters can be optimized to make the output results more accurate, and horizontally, more generalized or scenario-specific factors can be incorporated to enrich the judgment basis and improve the robustness.

[0075] Based on the above technical solution, referring to a collision detection processing flow chart as shown in Figure 4 : It includes the following steps: Obstacle trailer classification - input the detected trailers of other vehicles and the self-vehicle trailers generated due to misdetection into the recognition algorithm, respectively perform the recognition of the trailers of other vehicles based on the target tractor and the recognition of the trailers of other vehicles based on the target trailer, finally obtain the recognition result of the trailers of other vehicles, input this recognition result into the collision detection algorithm for processing, and then perform the planning of vehicle behavior.

[0076] Figure 5 It is a structural schematic diagram of an obstacle recognition device in an embodiment of the present disclosure. As shown in Figure 5 : The device includes: an acquisition module 510, a first determination module 520, and a second determination module 530.

[0077] An acquisition module 510, configured to acquire trailer information of a detected trailer within the perception range of a target vehicle, where the target vehicle includes a target tractor and a target trailer towed by the target tractor; a first determination module 520, configured to determine the trailer type of the detected trailer according to the trailer information and the associated information of the target tractor, and / or determine the trailer type of the detected trailer according to the trailer information and the associated information of the target trailer; a second determination module 530, configured to determine whether the detected trailer is an obstacle affecting the driving of the target vehicle according to the trailer type of the detected trailer.

[0078] The first determination module 520 includes: a division unit, configured to divide the target tractor into a first part including the vehicle head and a second part including the vehicle tail according to the associated information of the target tractor; a first determination unit, configured to determine the trailer type of the detected trailer based on the first part and / or the second part and the trailer information.

[0079] The trailer type includes a trailer of another vehicle and a self-trailer; the first determination unit is specifically configured to: determine whether the detected trailer has a collision risk with the first part and whether it has a collision risk with the second part based on the trailer information; if the detected trailer has a collision risk with the first part, determine the detected trailer as a trailer of another vehicle; if the detected trailer has a collision risk with the second part, determine the detected trailer as a self-trailer; correspondingly, the second determination module 530 is configured to, if the detected trailer is a trailer of another vehicle, determine the detected trailer as an obstacle affecting the driving of the target vehicle.

[0080] The division unit is specifically configured to: determine a corresponding rectangle according to the width and length of the target tractor, and determine the four vertex coordinates of the rectangle according to the position information in the associated information of the target tractor; determine the area range of the first part and the area range of the second part according to the segmentation coefficient and the four vertex coordinates of the rectangle; where the segmentation coefficient is determined according to the width, length, center point of the target tractor, and the possible angle range of the obstacle.

[0081] Optionally, the first determination module 520 includes: a second determination unit, configured to determine one or more of the morphological features, size features, and motion features of the detected trailer according to the trailer information; and determine the trailer type of the detected trailer according to one or more of the morphological features, size features, and motion features and the associated information of the target trailer.

[0082] Optionally, the morphological feature includes the area of the detected trailer, the dimensional feature includes the maximum side length of the detected trailer, and the motion feature includes the magnitude and direction of the speed of the detected trailer; the associated information of the target trailer includes the area of the target trailer, the maximum side length of the target trailer, the magnitude of the speed of the target trailer, the direction of the speed of the target trailer, and the appearance area of the target trailer; the second determination unit includes: a construction subunit configured to construct an influencing factor according to one or more of the morphological feature, the dimensional feature, and the motion feature, and the associated information of the target trailer; a determination subunit configured to determine the trailer type of the detected trailer according to the influencing factor; wherein, the influencing factor includes any one or more of the following: a first ratio of the area of the detected trailer to the area of the target trailer, a second ratio of the maximum side length of the detected trailer to the maximum side length of the target trailer, a third ratio of the magnitude of the speed of the detected trailer to the magnitude of the speed of the target trailer, a fourth ratio of the direction of the speed of the detected trailer to the direction of the speed of the target trailer, and result information on whether the geometric center of the detected trailer is a point within the appearance area of the target trailer.

[0083] Optionally, the determination subunit is specifically configured to: determine first data for characterizing the trailer type of the detected trailer according to the first ratio; determine second data for characterizing the trailer type of the detected trailer according to the second ratio; determine third data for characterizing the trailer type of the detected trailer according to the third ratio; determine fourth data for characterizing the trailer type of the detected trailer according to the fourth ratio; determine fifth data for characterizing the trailer type of the detected trailer according to the result information on whether the geometric center of the detected trailer is a point within the appearance area of the target trailer; perform weighted summation on the first data, the second data, the third data, the fourth data, and the fifth data according to weights respectively matching the first ratio, the second ratio, the third ratio, the fourth ratio, and the result information to obtain a reference value; and determine the trailer type of the detected trailer according to the parameter value and a preset threshold.

[0084] The obstacle recognition device provided in an embodiment of the present disclosure can execute the steps in the obstacle recognition method provided in the method embodiment of the present disclosure, and has corresponding functional modules and beneficial effects for executing the method.

[0085] Figure 6 It is a schematic structural diagram of an electronic device in an embodiment of the present disclosure. Specifically, refer to Figure 6 below, which shows a schematic structural diagram of an electronic device 500 suitable for implementing the present disclosure. Figure 6 The electronic device shown is only an example and should not impose any limitation on the functions and usage scope of the embodiments of the present disclosure.

[0086] As Figure 6 shown, the electronic device 500 may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 501, which may perform various appropriate actions and processes according to a program stored in the read-only memory (ROM) 502 or a program loaded from the storage device 508 into the random access memory (RAM) 503 to implement the method of the embodiments described in the present disclosure. In the RAM 503, various programs and data required for the operation of the electronic device 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. The input / output (I / O) interface 505 is also connected to the bus 504.

[0087] Specifically, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, the embodiments of the present disclosure include a computer program product that includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes program codes for performing the methods shown in the flowcharts, thereby implementing the obstacle recognition method as described above. In such an embodiment, the computer program may be downloaded and installed from the network through the communication device 509, or installed from the storage device 508, or installed from the ROM 502. When the computer program is executed by the processing device 501, the above-mentioned functions defined in the method of the embodiments of the present disclosure are executed.

[0088] It should be noted that the above-mentioned computer-readable medium in the present disclosure can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. In the present disclosure, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, and this computer-readable signal medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0089] The above-mentioned computer-readable medium can be included in the above-mentioned electronic device; it can also exist separately without being assembled into the electronic device. The above-mentioned computer-readable medium carries one or more programs, and when the above-mentioned one or more programs are executed by the electronic device, the electronic device is caused to execute the above-mentioned obstacle recognition method.

[0090] Optionally, when the above-mentioned one or more programs are executed by the electronic device, the electronic device can also execute the other steps described in the above-mentioned embodiments.

[0091] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0092] The above description is only a preferred embodiment of the present disclosure and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosure concept. For example, the technical solutions formed by mutually replacing the above features with the (but not limited to) technical features having similar functions disclosed in the present disclosure.

Claims

1. An obstacle recognition method, characterized in that, The method includes: Obtaining trailer information of a detected trailer within the perception range of a target vehicle, where the target vehicle includes a target tractor and a target trailer towed by the target tractor; Dividing the target tractor into a first part including the front of the vehicle and a second part including the rear of the vehicle according to the associated information of the target tractor, determining whether there is a collision risk between the detected trailer and the first part and whether there is a collision risk between the detected trailer and the second part based on the trailer information; if there is a collision risk between the detected trailer and the first part, determining the detected trailer as a trailer of another vehicle; if there is a collision risk between the detected trailer and the second part, determining the detected trailer as the self - vehicle trailer; if the detected trailer is a trailer of another vehicle, determining the detected trailer as an obstacle affecting the driving of the target vehicle; And / or, determining the trailer type of the detected trailer according to the trailer information and the associated information of the target trailer; determining whether the detected trailer is an obstacle affecting the driving of the target vehicle according to the trailer type of the detected trailer; the determining the trailer type of the detected trailer according to the trailer information and the associated information of the target trailer includes: determining one or more of the morphological features, size features, and motion features of the detected trailer according to the trailer information; determining the trailer type of the detected trailer according to one or more of the morphological features, size features, and motion features and the associated information of the target trailer.

2. The method according to claim 1, wherein The dividing the target tractor into a first part including the front of the vehicle and a second part including the rear of the vehicle according to the associated information of the target tractor includes: Determining a corresponding rectangle according to the width and length of the target tractor, and determining the four vertex coordinates of the rectangle according to the position information in the associated information of the target tractor; Determining the area range of the first part and the area range of the second part according to the segmentation coefficient and the four vertex coordinates of the rectangle; Wherein, the segmentation coefficient is determined according to the width, length, center point of the target tractor, and the possible angle range of the obstacle.

3. The method according to claim 1, characterized in that, The morphological features include the area of the detected trailer, the size features include the maximum side length of the detected trailer, and the motion features include the speed magnitude and speed direction of the detected trailer; The associated information of the target trailer includes the area of the target trailer, the maximum side length of the target trailer, the speed magnitude of the target trailer, the speed direction of the target trailer, and the appearance area of the target trailer; Correspondingly, the determining the trailer type of the detected trailer according to one or more of the morphological features, size features, and motion features and the associated information of the target trailer includes: Constructing influencing factors according to one or more of the morphological features, size features, and motion features and the associated information of the target trailer; Determining the trailer type of the detected trailer according to the influencing factors; Among them, the influencing factors include any one or more of the following: the first ratio of the area of the detected trailer to the area of the target trailer, the second ratio of the maximum side length of the detected trailer to the maximum side length of the target trailer, the third ratio of the speed magnitude of the detected trailer to the speed magnitude of the target trailer, the fourth ratio of the speed direction of the detected trailer to the speed direction of the target trailer, and the result information on whether the geometric center of the detected trailer is a point within the appearance area of the target trailer.

4. The method according to claim 3, characterized in that, Determining the trailer type of the detected trailer according to the influencing factors includes: Determining first data for characterizing the trailer type of the detected trailer according to the first ratio; Determining second data for characterizing the trailer type of the detected trailer according to the second ratio; Determining third data for characterizing the trailer type of the detected trailer according to the third ratio; Determining fourth data for characterizing the trailer type of the detected trailer according to the fourth ratio; Determining fifth data for characterizing the trailer type of the detected trailer according to the result information on whether the geometric center of the detected trailer is a point within the appearance area of the target trailer; Performing weighted summation on the first data, the second data, the third data, the fourth data, and the fifth data according to weights respectively matching the first ratio, the second ratio, the third ratio, the fourth ratio, and the result information to obtain a reference value; Determining the trailer type of the detected trailer according to the reference value and a preset threshold.

5. An obstacle recognition device, characterized in that, Including: An acquisition module, configured to acquire trailer information of a detected trailer within the perception range of a target vehicle, where the target vehicle includes a target tractor and a target trailer towed by the target tractor; A first determination module, configured to divide the target tractor into a first part including the vehicle head and a second part including the vehicle tail according to the associated information of the target tractor; determine whether the detected trailer has a collision risk with the first part and whether it has a collision risk with the second part based on the trailer information; if the detected trailer has a collision risk with the first part, determine the detected trailer as a trailer of another vehicle; if the detected trailer has a collision risk with the second part, determine the detected trailer as a self-owned trailer, and / or determine the trailer type of the detected trailer according to the trailer information and the associated information of the target trailer; A second determination module, configured to determine whether the detected trailer is an obstacle affecting the driving of the target vehicle according to the trailer type of the detected trailer; wherein, if the detected trailer is a trailer of another vehicle, determine the detected trailer as an obstacle affecting the driving of the target vehicle; The determining the trailer type of the detected trailer according to the trailer information and the associated information of the target trailer includes: determining one or more of the morphological characteristics, size characteristics, and motion characteristics of the detected trailer according to the trailer information; determining the trailer type of the detected trailer according to one or more of the morphological characteristics, size characteristics, and motion characteristics, and the associated information of the target trailer.

6. An electronic device, characterized in that, The electronic device includes: One or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, the method according to any one of claims 1-4 is implemented.

Images