Blind spot warning method and system for heavy-duty semi-trailer tractor

Abstract

The present invention provides a blind spot warning method and system for a heavy-duty semi-trailer tractor. Based on the structural information of the heavy-duty semi-trailer tractor, factors influencing the formation of the blind spot of the tractor are determined, and the correlation information between the turning radius and the visual blind spot range of the tractor in different turning scenes is fitted and generated, and the changes in the visual blind spot range of the tractor under different turning actions are continuously marked; the road ahead of the tractor is visually identified, and the characteristic information of the turning action that will occur in the future driving process of the tractor is predicted, and the actual visual blind spot azimuth state information of the tractor in the future driving process is determined in combination with the correlation information, and the visual blind spot of the tractor is accurately predicted to provide guidance for subsequent radar detection; based on the actual visual blind spot range state information, the radar equipment on the tractor is instructed to perform detection, and the existence state information of the objects in the actual visual blind spot range is obtained, so as to generate a driving notification message and / or an external alarm reminder message, and the radar detection method is used as an auxiliary detection method for the visual blind spot of the tractor to achieve global effective detection of the visual blind spot.

Description

Technical Field

[0001] The present invention relates to the field of vehicle control, and in particular to a blind spot warning method and system for a heavy-duty semi-trailer tractor. Background Art

[0002] As a heavy-duty engineering vehicle, a heavy-duty semi-trailer tractor has a long wheelbase, which results in a large inner wheel difference when the tractor is turning, causing the driver to have a visual blind spot for pedestrians or vehicles and other objects within the inner wheel difference coverage area, and unable to observe the objects within the inner wheel difference coverage area in a timely and accurate manner. In order to overcome this defect of heavy-duty semi-trailer tractors, rearview mirrors are installed on the tractors. Although rearview mirrors can increase the driver's field of vision to a certain extent, they cannot effectively improve the driver's viewing field for the huge visual blind spot of heavy-duty semi-trailer tractors. In addition, when the tractor makes continuous turns, the area covered by its inner wheel difference also changes continuously accordingly, which adds more uncertainty to the changes in the driver's field of vision blind spot, thereby reducing the driving safety of heavy-duty semi-trailer tractors. Summary of the invention

[0003] The object of the present invention is to provide a blind spot warning method and system for a heavy-duty semi-trailer tractor. The method and system determine the factors affecting the formation of the blind spot of the tractor based on the structural information of the heavy-duty semi-trailer tractor, thereby fitting and generating the correlation information between the turning radius and the visual blind spot range of the tractor in different turning scenes, and continuously marking the changes in the visual blind spot range of the tractor under different turning actions; visually identify the road ahead of the tractor, predict the turning action feature information of the tractor in the future driving process, and determine the actual visual blind spot azimuth state information of the tractor in the future driving process in combination with the correlation information, accurately predict the visual blind spot of the tractor, and provide guidance for subsequent radar detection; then, based on the actual visual blind spot range state information, instruct the radar equipment on the tractor to perform detection, obtain the object existence state information in the actual visual blind spot range, and generate a driving notification message and / or an external alarm reminder message, and use the radar detection method as an auxiliary detection method for the visual blind spot of the tractor to achieve global effective detection of the visual blind spot and improve the driving safety of the tractor.

[0004] The present invention is achieved through the following technical solutions:

[0005] The blind spot warning method for heavy-duty semi-trailer tractors includes:

[0006] Based on the structural information of the heavy-duty semi-trailer tractor, determining the factors affecting the formation of the blind spot of the tractor; based on the factors affecting the formation of the blind spot, fitting and generating the correlation information between the turning radius and the visual blind spot range of the tractor in different turning scenarios;

[0007] Acquire a road image ahead of the tractor, analyze the road image ahead, and predict characteristic information of turning actions of the tractor in the future driving process; determine actual visual blind spot range state information of the tractor in the future driving process based on the turning action characteristic information and the associated information;

[0008] Based on the actual visual blind spot range status information, instruct the radar equipment on the tractor to detect the corresponding surrounding area of ​​the tractor, so as to obtain the object existence status information within the actual visual blind spot range; based on the object existence status information, generate corresponding driving notification messages and / or external vehicle alarm reminder messages.

[0009] Optionally, based on the structural information of the heavy-duty semi-trailer tractor, determining the factors affecting the formation of the blind spot of the tractor; and based on the factors affecting the formation of the blind spot, fitting and generating the correlation information between the turning radius and the visual blind spot range of the tractor in different turning scenarios, including:

[0010] Based on the length dimension information of the heavy semi-trailer tractor, the inner wheel difference of the tractor is determined, and based on the shape dimension information of the rearview mirror of the tractor, the field of view range of the rearview mirror of the tractor is determined, so that the inner wheel difference and the field of view range of the rearview mirror are determined as the influencing factors of the blind spot formation of the tractor;

[0011] Based on the inner wheel difference and the field of view of the rearview mirror, the visual blind area ranges corresponding to the tractor in several turning radius scenarios are determined; wherein the several turning radius scenarios correspond to several discrete turning radii with the same radius difference; neural network learning is performed based on all the visual blind area ranges corresponding to the several turning radius scenarios, and the correlation information between the turning radius of the tractor and the size and distribution position of the visual blind area range in different turning scenarios where the turning radius continuously changes is fitted and generated.

[0012] Optionally, obtaining a road image ahead of the tractor, analyzing the road image ahead, and predicting characteristic information of turning actions that may occur in the future driving process of the tractor; and determining actual visual blind spot range state information of the tractor in the future driving process based on the turning action characteristic information and the associated information, including:

[0013] Scan and photograph the road ahead of the tractor to obtain a panoramic image of the road ahead; perform road contour recognition processing on the panoramic image of the road ahead to obtain extended shape information of the road ahead of the tractor; based on the extended shape information of the road and the driving speed of the tractor, predict the change information of the turning radius corresponding to the turning action of the tractor in the future driving process, and use this as the turning action feature information;

[0014] The turning radius change information is compared and analyzed with the associated information to determine the actual visual blind spot range size change information and distribution position change information corresponding to the turning action of the tractor in the future driving process, and use this as the actual visual blind spot range state information.

[0015] Optionally, based on the actual visual blind spot range status information, instructing the radar device on the tractor to detect the corresponding surrounding area of ​​the tractor, so as to obtain the object existence status information within the actual visual blind spot range; based on the object existence status information, generating a corresponding driving notification message and / or an external vehicle alarm reminder message, including:

[0016] Based on the actual visual blind area size change information and distribution position change information included in the actual visual blind area state information, determine the occupied area range of the actual visual blind area in the three-dimensional space where the tractor is located; based on the occupied area range, instruct the radar equipment on the tractor to track and detect the occupied area range to obtain corresponding radar detection data; then analyze the radar detection data to obtain the object existence state information in the actual visual blind area; wherein the object existence state information includes the relative distance and relative orientation information between the object and the tractor;

[0017] Based on the object existence status information, it is determined whether the tractor will collide with the corresponding object; when a collision occurs, a driving notification message including a driving speed reduction and / or an external vehicle alarm reminder message including a preset voice signal is generated.

[0018] Optionally, based on the occupied area range, instructing a radar device on the tractor to track and detect the occupied area range, including:

[0019] Step S1, establish a spatial rectangular coordinate system in the occupied area, so that the points inside it can be expressed by coordinates, and use the following formula (1) to determine the scanning range of the radar device on the tractor according to the occupied area:

[0020] R=1.5×max a,b∈Q {L[(x a ,y a ,z a ),(x b ,y b ,z b )]} (1)

[0021] In the above formula (1), R indicates that the scanning range of the radar device on the tractor is a spherical area with a radius of R; a,b∈Q indicates that both position a and position b are points within the occupied area; (xa ,y a ,z a ) represents the coordinate point of position a within the occupied area; (x b ,y b ,z b ) represents the coordinate point b within the occupied area; L[(x a ,y a ,z a ),(x b ,y b ,z b )] indicates the spatial distance between the coordinate points of position a and position b; max a,b∈Q {L[(x a ,y a ,z a ),(x b ,y b ,z b )]} means to obtain the maximum value of the spatial distance between the coordinate points of position a and position b under the condition that a, b∈Q is satisfied;

[0022] Step S2, using the following formula (2), according to the scanning range of the radar device on the tractor and the single-point scanning range of the radar device, determine the mobile scanning mode of the radar device,

[0023]

[0024] In the above formula (2), Indicates that the radar device moves from the current radar device position (x0, y0, z0) to the position point Perform reciprocating movement of L distance for mobile scanning, where L = (R-r0); r0 represents the single-point scanning radius of the radar device on the tractor;

[0025] Step S3, using the following formula (3), according to the scanning range of the radar device on the tractor and the single-point scanning range of the radar device, determine the moving speed of the radar device during the mobile scanning.

[0026]

[0027] In the above formula (3), V represents the moving speed of the radar device during the moving scan; V0 represents the minimum moving speed of the radar device during the moving scan; V M Indicates the maximum moving speed of the radar device when moving and scanning.

[0028] Heavy-duty semi-trailer tractor blind spot warning system, including:

[0029] A blind spot formation influencing factor determination module, used to determine the blind spot formation influencing factors of the tractor based on the structural information of the heavy-duty semi-trailer tractor;

[0030] A turning radius-blind spot association identification module, used for fitting and generating association information between the turning radius and the visual blind spot range of the tractor in different turning scenarios based on the influencing factors of the blind spot formation;

[0031] A turning action feature recognition module is used to obtain a road image ahead of the tractor, analyze the road image ahead, and predict turning action feature information that may occur during the tractor's future driving process;

[0032] An actual visual blind area determination module, used to determine the actual visual blind area range state information of the tractor during future driving based on the turning action feature information and the associated information;

[0033] A radar detection control module, for instructing the radar device on the tractor to detect the corresponding surrounding area of ​​the tractor based on the actual visual blind area range status information, so as to obtain the object existence status information within the actual visual blind area range;

[0034] The early warning notification module is used to generate corresponding driving notification messages and / or external vehicle alarm reminder messages based on the object existence status information.

[0035] Optionally, the blind spot formation influencing factor determination module is used to determine the blind spot formation influencing factors of the tractor based on the structural information of the heavy-duty semi-trailer tractor, including:

[0036] Based on the length dimension information of the heavy semi-trailer tractor, the inner wheel difference of the tractor is determined, and based on the shape dimension information of the rearview mirror of the tractor, the field of view range of the rearview mirror of the tractor is determined, so that the inner wheel difference and the field of view range of the rearview mirror are determined as the influencing factors of the blind spot formation of the tractor;

[0037] The turning radius-blind spot association identification module is used to fit and generate association information between the turning radius and the visual blind spot range of the tractor in different turning scenarios based on the influencing factors of the blind spot formation, including:

[0038] Based on the inner wheel difference and the field of view of the rearview mirror, the visual blind area ranges corresponding to the tractor in several turning radius scenarios are determined; wherein the several turning radius scenarios correspond to several discrete turning radii with the same radius difference; neural network learning is performed based on all the visual blind area ranges corresponding to the several turning radius scenarios, and the correlation information between the turning radius of the tractor and the size and distribution position of the visual blind area range in different turning scenarios where the turning radius continuously changes is fitted and generated.

[0039] Optionally, the turning action feature recognition module is used to obtain a road image ahead of the tractor, analyze the road image ahead, and predict turning action feature information of the tractor during future driving, including:

[0040] Scan and photograph the road ahead of the tractor to obtain a panoramic image of the road ahead; perform road contour recognition processing on the panoramic image of the road ahead to obtain extended shape information of the road ahead of the tractor; based on the extended shape information of the road and the driving speed of the tractor, predict the change information of the turning radius corresponding to the turning action of the tractor in the future driving process, and use this as the turning action feature information;

[0041] The actual visual blind area determination module is used to determine the actual visual blind area range state information of the tractor during future driving based on the turning action feature information and the associated information, including:

[0042] The turning radius change information is compared and analyzed with the associated information to determine the actual visual blind spot range size change information and distribution position change information corresponding to the turning action of the tractor in the future driving process, and use this as the actual visual blind spot range state information.

[0043] Optionally, the radar detection control module is used to instruct the radar device on the tractor to detect the corresponding surrounding area of ​​the tractor based on the actual visual blind area range status information, so as to obtain the object existence status information within the actual visual blind area range, including:

[0044] Based on the actual visual blind area size change information and distribution position change information included in the actual visual blind area state information, determine the occupied area range of the actual visual blind area in the three-dimensional space where the tractor is located; based on the occupied area range, instruct the radar equipment on the tractor to track and detect the occupied area range to obtain corresponding radar detection data; then analyze the radar detection data to obtain the object existence state information in the actual visual blind area; wherein the object existence state information includes the relative distance and relative orientation information between the object and the tractor;

[0045] The early warning notification module is used to generate a corresponding driving notification message and / or an external vehicle alarm reminder message based on the object existence status information, including:

[0046] Based on the object existence status information, it is determined whether the tractor will collide with the corresponding object; when a collision occurs, a driving notification message including a driving speed reduction and / or an external vehicle alarm reminder message including a preset voice signal is generated.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] The blind spot warning method and system for a heavy-duty semi-trailer tractor provided in the present application are based on the structural information of the heavy-duty semi-trailer tractor, determine the factors affecting the formation of the blind spot of the tractor, and thereby fit and generate the correlation information between the turning radius and the visual blind spot range of the tractor in different turning scenarios, and continuously mark the changes in the visual blind spot range of the tractor under different turning actions; visually identify the road ahead of the tractor, predict the characteristic information of the turning action that will occur in the future driving process of the tractor, and determine the actual visual blind spot azimuth state information of the tractor in the future driving process in combination with the correlation information, accurately predict the visual blind spot of the tractor, and provide guidance for subsequent radar detection; then, based on the actual visual blind spot range state information, instruct the radar equipment on the tractor to perform detection, and obtain the object existence state information within the actual visual blind spot range, so as to generate a driving notification message and / or an external alarm reminder message, and use the radar detection method as an auxiliary detection method for the visual blind spot of the tractor to achieve global and effective detection of the visual blind spot, thereby improving the driving safety of the tractor. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0050] Figure 1 A schematic flow chart of a blind spot warning method for a heavy-duty semi-trailer tractor provided by the present invention.

[0051] Figure 2 This is a structural schematic diagram of the blind spot warning system for the heavy-duty semi-trailer tractor provided by the present invention. DETAILED DESCRIPTION

[0052] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. It is to be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some structures related to the present application are shown in the accompanying drawings, rather than all structures. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0053] The terms "including" and "having" and any variations thereof in this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0054] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0055] See also Figure 1 As shown, an embodiment of the present application provides a blind spot warning method for a heavy-duty semi-trailer tractor. The blind spot warning method for a heavy-duty semi-trailer tractor comprises:

[0056] Based on the structural information of the heavy-duty semi-trailer tractor, determine the factors affecting the formation of the blind spot of the tractor; based on the factors affecting the formation of the blind spot, fit and generate the correlation information between the turning radius and the visual blind spot range of the tractor in different turning scenarios;

[0057] Acquire a road image ahead of the tractor, analyze the road image ahead, and predict characteristic information of turning actions that may occur in the tractor during future driving; determine actual visual blind spot range state information of the tractor during future driving based on the turning action characteristic information and the associated information;

[0058] Based on the actual visual blind spot range status information, the radar equipment on the tractor is instructed to detect the corresponding surrounding area of ​​the tractor, so as to obtain the object existence status information within the actual visual blind spot range; based on the object existence status information, a corresponding driving notification message and / or an external vehicle alarm reminder message is generated.

[0059] The beneficial effects of the above embodiments are as follows: the blind spot warning method for a heavy-duty semi-trailer tractor determines the factors affecting the formation of the blind spot of the tractor based on the structural information of the heavy-duty semi-trailer tractor, thereby fitting and generating the correlation information between the turning radius and the visual blind spot range of the tractor in different turning scenarios, and continuously marking the changes in the visual blind spot range of the tractor under different turning actions; visually identifying the road ahead of the tractor, predicting the turning action feature information of the tractor in the future driving process, and combining the correlation information to determine the actual visual blind spot orientation state information of the tractor in the future driving process, accurately predicting the visual blind spot of the tractor, and providing guidance for subsequent radar detection; then based on the actual visual blind spot range state information, instructing the radar equipment on the tractor to perform detection, and obtaining the object existence state information within the actual visual blind spot range, thereby generating a driving notification message and / or an external alarm reminder message, and using the radar detection method as an auxiliary detection method for the visual blind spot of the tractor to achieve global effective detection of the visual blind spot, thereby improving the driving safety of the tractor.

[0060] In another embodiment, based on the structural information of the heavy-duty semi-trailer tractor, factors influencing the formation of the blind spot of the tractor are determined; based on the factors influencing the formation of the blind spot, association information between the turning radius and the visual blind spot range of the tractor in different turning scenarios is fitted and generated, including:

[0061] Based on the length dimension information of the heavy semi-trailer tractor, the inner wheel difference of the tractor is determined, and based on the shape dimension information of the rearview mirror of the tractor, the field of view range of the rearview mirror of the tractor is determined, so that the inner wheel difference and the field of view range of the rearview mirror are determined as the influencing factors of the blind spot formation of the tractor;

[0062] Based on the inner wheel difference and the field of view of the rearview mirror, the visual blind area ranges corresponding to the tractor in several turning radius scenarios are determined; wherein the several turning radius scenarios correspond to several discrete turning radii with the same radius difference; neural network learning is performed based on all the visual blind area ranges corresponding to the several turning radius scenarios, and the correlation information between the turning radius of the tractor and the size and distribution position of the visual blind area range in different turning scenarios in which the turning radius continuously changes is fitted and generated.

[0063] The beneficial effect of the above embodiment is that the inner wheel difference of the heavy semi-trailer tractor is the determining factor of the visual blind spot generated by the tractor during the turning process. The larger the inner wheel difference, the larger the corresponding visual blind spot range. The inner wheel difference of the tractor is positively correlated with the length dimension information of the tractor (such as the wheelbase length). By calculating the wheelbase length of the tractor, the inner wheel difference of the tractor can be accurately obtained. In addition, the rearview mirror is set in the tractor to expand the driver's field of view, and the field of view expansion effect of rearview mirrors of different shapes and sizes is also different. Similarly, when the mirror area of ​​the rearview mirror is larger, the field of view expansion rate is also larger; when the rearview mirror is a convex mirror and the radius of curvature of the rearview mirror is smaller, the field of view expansion rate is also larger; when the field of view expansion rate of the rearview mirror is larger, the corresponding blind spot range is also smaller. Therefore, the inner wheel difference of the tractor and the field of view of the rearview mirror are used as the influencing factors of the blind spot formation of the tractor, which can provide a reliable basis for the subsequent calculation of the size of the visual blind spot formed by the tractor during the turning process. Based on the inner wheel difference and the field of view of the rearview mirror, the visual blind area range corresponding to each of the tractor in several turning radius scenarios is determined, that is, based on several discrete turning radii, the size and distribution position of the visual blind area range of the tractor in the corresponding turning process are calculated, so as to generate the basic data for the subsequent fitting of the size and distribution position state of the visual blind area range of the tractor in the dynamic turning process; wherein, the several discrete turning radii can be but not limited to several turning radii with the same radius difference. Based on all the visual blind area ranges corresponding to the several turning radius scenarios, neural network learning is performed to fit and generate the correlation information between the turning radius and the size and distribution position of the visual blind area range of the tractor in different turning scenarios with continuously changing turning radius; wherein the neural network learning can be implemented by using a convolutional neural network model, which belongs to the conventional technical means in the field and will not be described in detail here. Through the above method, the correlation between the turning radius of the tractor in different turning scenarios with continuously changing turning radius and the size and distribution position of the visual blind area range can be quantitatively characterized, so as to facilitate the subsequent rapid and accurate determination of the visual blind area range formed in real time according to the actual turning situation of the tractor.

[0064] In another embodiment, an image of the road ahead of the tractor is obtained, the image of the road ahead is analyzed, and characteristic information of a turning action of the tractor in the future driving process is predicted; based on the characteristic information of the turning action and the associated information, actual visual blind spot range state information of the tractor in the future driving process is determined, including:

[0065] Scan and photograph the road ahead of the tractor to obtain a panoramic image of the road ahead; perform road contour recognition processing on the panoramic image of the road ahead to obtain extended shape information of the road ahead of the tractor; based on the extended shape information of the road and the driving speed of the tractor, predict the change information of the turning radius corresponding to the turning action of the tractor in the future driving process, and use this as the characteristic information of the turning action;

[0066] The turning radius change information is compared and analyzed with the associated information to determine the actual visual blind spot range size change information and distribution position change information corresponding to the turning action of the tractor in the future driving process, and use this as the actual visual blind spot range state information.

[0067] The beneficial effect of the above embodiment is that by scanning and photographing the road ahead of the tractor, a corresponding panoramic image of the road ahead is obtained, so that the shape of the road ahead that the tractor may pass through during the driving process can be globally recognized, and then the panoramic image of the road ahead is processed to obtain the extended shape information of the road ahead of the tractor (such as the curvature of the road, etc.), and combined with the driving speed of the tractor, the corresponding turning radius change information of the turning action of the tractor in the future driving process is predicted, and this is used as the characteristic information of the turning action. Since the inertia of the tractor is large and the center of gravity is high, in order to avoid the tractor from tilting during the turning process, the driver usually actively reduces the turning speed of the tractor, and the turning speed of the tractor will affect the turning radius of the tractor itself. According to the centrifugal principle of the tractor during the turning process, the turning speed of the tractor is inversely proportional to the turning radius. By combining the road extension shape information and the driving speed of the tractor, the turning radius change of the tractor during the turning process can be accurately predicted and calculated, providing a reliable basis for the subsequent dynamic determination of the actual visual blind spot range of the tractor. The turning radius change information is compared and analyzed with the associated information to determine the actual visual blind spot range size change information and distribution position change information corresponding to the tractor turning action in the future driving process, so as to accurately identify the visual blind spot situation of the tractor during the entire turning process.

[0068] In another embodiment, based on the actual visual blind spot range state information, instructing the radar device on the tractor to detect the corresponding surrounding area of ​​the tractor, so as to obtain the object existence state information within the actual visual blind spot range; based on the object existence state information, generating a corresponding driving notification message and / or an external vehicle alarm reminder message, including:

[0069] Based on the actual visual blind area size change information and distribution position change information included in the actual visual blind area state information, determine the occupied area range of the actual visual blind area in the three-dimensional space where the tractor is located; based on the occupied area range, instruct the radar equipment on the tractor to track and detect the occupied area range to obtain corresponding radar detection data; and then analyze the radar detection data to obtain the object existence state information in the actual visual blind area; wherein the object existence state information includes the relative distance and relative orientation information between the object and the tractor;

[0070] Based on the object existence status information, it is determined whether the tractor will collide with the corresponding object; when a collision occurs, a driving notification message including a driving speed reduction and / or an external vehicle alarm reminder message including a preset voice signal is generated.

[0071] The beneficial effect of the above embodiment is that the actual visual blind area is an area that the driver on the tractor cannot see through the naked eye or the rearview mirror. In order to compensate for the invisibility of the actual visual blind area, the actual visual blind area size change information and distribution position change information contained in the actual visual blind area state information are used as a reference to determine the occupied area of ​​the actual visual blind area in the three-dimensional space where the tractor is located (that is, the occupied area formed by the corresponding projection of the actual visual blind area in the three-dimensional space), so as to perform a three-dimensional representation of the actual visual blind area. Based on the occupied area, the radar device on the tractor is instructed to track and detect the occupied area, that is, the radar device is instructed to perform real-time aiming detection on the occupied area to obtain corresponding radar detection data, so as to calculate the existence status information of the object in the actual visual blind area, so as to accurately locate and identify the objects in the area around the tractor. In addition, based on the existence status information of the object, it is determined whether the distance between the tractor and the object is less than a preset distance threshold. If so, it is determined that the tractor will collide with the corresponding object. If not, it is determined that the tractor will not collide with the corresponding object, and when a collision occurs, a driving notification message including a driving speed reduction and / or an external vehicle alarm reminder message including a preset voice signal is generated, so that the driver of the tractor and the personnel located in the actual visual blind spot of the tractor can be reminded and notified in a timely and accurate manner.

[0072] In another embodiment, based on the occupied area range, instructing the radar device on the tractor to track and detect the occupied area range, including:

[0073] Step S1, establish a spatial rectangular coordinate system within the occupied area, and construct a corresponding spatial rectangular coordinate system with the radar device as the origin, so that the points inside it can be expressed by coordinates. The following formula (1) is used to determine the scanning range of the radar device on the tractor according to the occupied area:

[0074] R=1.5×max a,b∈Q {L[(x a ,y a ,z a ),(x b ,y b ,z b )]} (1)

[0075] In the above formula (1), R indicates that the scanning range of the radar device on the tractor is a spherical area with a radius of R; a,b∈Q indicates that both position a and position b are points within the occupied area; (x a ,y a ,z a ) represents the coordinate point a within the occupied area; (x b ,y b ,z b ) represents the coordinate point b within the occupied area; L[(x a ,y a ,z a ),(x b ,y b ,z b )] indicates the spatial distance between the coordinate points of position a and position b; max a,b∈Q {L[(x a ,y a ,z a ),(x b ,y b ,z b )]} means to obtain the maximum value of the spatial distance between the coordinate points of position a and position b under the condition that a, b∈Q is satisfied;

[0076] Step S2, using the following formula (2), according to the scanning range of the radar device on the tractor and the single-point scanning range of the radar device, determine the mobile scanning mode of the radar device,

[0077]

[0078] In the above formula (2), Indicates that the radar device moves from the current radar device position (x0, y0, z0) to the position point Perform reciprocating movement L distance for mobile scanning, where L = (R-r0); r0 represents the single-point scanning radius of the radar device on the tractor;

[0079] Step S3, using the following formula (3), according to the scanning range of the radar device on the tractor and the single-point scanning range of the radar device, determine the moving speed of the radar device during the mobile scanning.

[0080]

[0081] In the above formula (3), V represents the moving speed of the radar device during the mobile scanning; V0 represents the minimum moving speed of the radar device during the mobile scanning; V M Indicates the maximum moving speed of the radar device during mobile scanning.

[0082] The beneficial effect of the above embodiment is that the scanning range of the radar device on the tractor is determined according to the occupied area by using the above formula (1), thereby ensuring that the occupied area can be scanned; then the mobile scanning mode of the radar device is determined according to the scanning range of the radar device on the tractor and the single-point scanning range of the radar device, thereby performing a global mobile scanning of the occupied area by mobile scanning, thereby ensuring the reliability of the system scanning; then the moving speed of the radar device during mobile scanning is determined according to the scanning range of the radar device on the tractor and the single-point scanning range of the radar device, thereby ensuring that the mobile scanning can be performed reliably and ensuring the stability of the system.

[0083] See also Figure 2 As shown, a heavy-duty semi-trailer tractor blind spot warning system provided in one embodiment of the present application. The heavy-duty semi-trailer tractor blind spot warning system includes:

[0084] A blind spot formation influencing factor determination module, used to determine the blind spot formation influencing factors of the tractor based on the structural information of the heavy-duty semi-trailer tractor;

[0085] A turning radius-blind spot association identification module is used to fit and generate association information between the turning radius and the visual blind spot range of the tractor in different turning scenarios based on the influencing factors of the blind spot formation;

[0086] A turning action feature recognition module is used to obtain the road image ahead of the tractor, analyze the road image ahead, and predict the turning action feature information that will occur during the future driving process of the tractor;

[0087] An actual visual blind area determination module, used to determine actual visual blind area range state information of the tractor during future driving based on the turning action feature information and the associated information;

[0088] A radar detection control module is used to instruct the radar device on the tractor to detect the corresponding surrounding area of ​​the tractor based on the actual visual blind area range status information, so as to obtain the object existence status information within the actual visual blind area range;

[0089] The early warning notification module is used to generate corresponding driving notification messages and / or external vehicle alarm reminder messages based on the object's existence status information.

[0090] The above-mentioned embodiments have the beneficial effects that the blind spot warning system for heavy-duty semi-trailer tractors determines the factors affecting the formation of the blind spots of the tractors based on the structural information of the heavy-duty semi-trailer tractors, thereby fitting and generating the correlation information between the turning radius and the visual blind spot range of the tractors in different turning scenarios, and continuously marking the changes in the visual blind spot range of the tractors under different turning actions; visually identifying the road ahead of the tractors, predicting the characteristic information of the turning actions that will occur in the future driving process of the tractors, and combining the correlation information to determine the actual visual blind spot orientation state information of the tractors in the future driving process, accurately predicting the visual blind spot of the tractors, and providing guidance for subsequent radar detection; and then instructing the radar equipment on the tractors to perform detection based on the actual visual blind spot range state information, and obtaining the object existence state information within the actual visual blind spot range, thereby generating a driving notification message and / or an external alarm reminder message, and using the radar detection method as an auxiliary detection method for the visual blind spot of the tractors, achieving global and effective detection of the visual blind spot, and improving the driving safety of the tractors.

[0091] In another embodiment, the blind spot formation influencing factor determination module is used to determine the blind spot formation influencing factors of the tractor based on the structural information of the heavy-duty semi-trailer tractor, including:

[0092] Based on the length dimension information of the heavy semi-trailer tractor, the inner wheel difference of the tractor is determined, and based on the shape dimension information of the rearview mirror of the tractor, the field of view range of the rearview mirror of the tractor is determined, so that the inner wheel difference and the field of view range of the rearview mirror are determined as the influencing factors of the blind spot formation of the tractor;

[0093] The turning radius-blind spot association identification module is used to fit and generate association information between the turning radius and the visual blind spot range of the tractor in different turning scenarios based on the influencing factors of the blind spot formation, including:

[0094] Based on the inner wheel difference and the field of view of the rearview mirror, the visual blind area ranges corresponding to the tractor in several turning radius scenarios are determined; wherein the several turning radius scenarios correspond to several discrete turning radii with the same radius difference; neural network learning is performed based on all the visual blind area ranges corresponding to the several turning radius scenarios, and the correlation information between the turning radius of the tractor and the size and distribution position of the visual blind area range in different turning scenarios in which the turning radius continuously changes is fitted and generated.

[0095] The beneficial effect of the above embodiment is that the inner wheel difference of the heavy semi-trailer tractor is the determining factor of the visual blind spot generated by the tractor during the turning process. The larger the inner wheel difference, the larger the corresponding visual blind spot range. The inner wheel difference of the tractor is positively correlated with the length dimension information of the tractor (such as the wheelbase length). By calculating the wheelbase length of the tractor, the inner wheel difference of the tractor can be accurately obtained. In addition, the rearview mirror is set in the tractor to expand the driver's field of view, and the field of view expansion effect of rearview mirrors of different shapes and sizes is also different. Similarly, when the mirror area of ​​the rearview mirror is larger, the field of view expansion rate is also larger; when the rearview mirror is a convex mirror and the radius of curvature of the rearview mirror is smaller, the field of view expansion rate is also larger; when the field of view expansion rate of the rearview mirror is larger, the corresponding blind spot range is also smaller. Therefore, the inner wheel difference of the tractor and the field of view of the rearview mirror are used as the influencing factors of the blind spot formation of the tractor, which can provide a reliable basis for the subsequent calculation of the size of the visual blind spot formed by the tractor during the turning process. Based on the inner wheel difference and the field of view of the rearview mirror, the visual blind area range corresponding to each of the tractor in several turning radius scenarios is determined, that is, based on several discrete turning radii, the size and distribution position of the visual blind area range of the tractor in the corresponding turning process are calculated, so as to generate the basic data for the subsequent fitting of the size and distribution position state of the visual blind area range of the tractor in the dynamic turning process; wherein, the several discrete turning radii can be but not limited to several turning radii with the same radius difference. Based on all the visual blind area ranges corresponding to the several turning radius scenarios, neural network learning is performed to fit and generate the correlation information between the turning radius and the size and distribution position of the visual blind area range of the tractor in different turning scenarios with continuously changing turning radius; wherein the neural network learning can be implemented by using a convolutional neural network model, which belongs to the conventional technical means in the field and will not be described in detail here. Through the above method, the correlation between the turning radius of the tractor in different turning scenarios with continuously changing turning radius and the size and distribution position of the visual blind area range can be quantitatively characterized, so as to facilitate the subsequent rapid and accurate determination of the visual blind area range formed in real time according to the actual turning situation of the tractor.

[0096] In another embodiment, the turning action feature recognition module is used to obtain the road image ahead of the tractor, analyze the road image ahead, and predict the turning action feature information of the tractor in the future driving process, including:

[0097] Scan and photograph the road ahead of the tractor to obtain a panoramic image of the road ahead; perform road contour recognition processing on the panoramic image of the road ahead to obtain extended shape information of the road ahead of the tractor; based on the extended shape information of the road and the driving speed of the tractor, predict the change information of the turning radius corresponding to the turning action of the tractor in the future driving process, and use this as the characteristic information of the turning action;

[0098] The actual visual blind area determination module is used to determine the actual visual blind area range state information of the tractor during future driving based on the turning action feature information and the associated information, including:

[0099] The turning radius change information is compared and analyzed with the associated information to determine the actual visual blind spot range size change information and distribution position change information corresponding to the turning action of the tractor in the future driving process, and use this as the actual visual blind spot range state information.

[0100] The beneficial effect of the above embodiment is that by scanning and photographing the road ahead of the tractor, a corresponding panoramic image of the road ahead is obtained, so that the shape of the road ahead that the tractor may pass through during the driving process can be globally recognized, and then the panoramic image of the road ahead is processed to obtain the extended shape information of the road ahead of the tractor (such as the curvature of the road, etc.), and combined with the driving speed of the tractor, the corresponding turning radius change information of the turning action of the tractor in the future driving process is predicted, and this is used as the characteristic information of the turning action. Since the inertia of the tractor is large and the center of gravity is high, in order to avoid the tractor from tilting during the turning process, the driver usually actively reduces the turning speed of the tractor, and the turning speed of the tractor will affect the turning radius of the tractor itself. According to the centrifugal principle of the tractor during the turning process, the turning speed of the tractor is inversely proportional to the turning radius. By combining the road extension shape information and the driving speed of the tractor, the turning radius change of the tractor during the turning process can be accurately predicted and calculated, providing a reliable basis for the subsequent dynamic determination of the actual visual blind spot range of the tractor. The turning radius change information is compared and analyzed with the associated information to determine the actual visual blind spot range size change information and distribution position change information corresponding to the tractor turning action in the future driving process, so as to accurately identify the visual blind spot situation of the tractor during the entire turning process.

[0101] In another embodiment, the radar detection control module is used to instruct the radar device on the tractor to detect the corresponding surrounding area of ​​the tractor based on the actual visual blind area range status information, so as to obtain the object existence status information within the actual visual blind area, including:

[0102] Based on the actual visual blind area size change information and distribution position change information included in the actual visual blind area state information, determine the occupied area range of the actual visual blind area in the three-dimensional space where the tractor is located; based on the occupied area range, instruct the radar equipment on the tractor to track and detect the occupied area range to obtain corresponding radar detection data; and then analyze the radar detection data to obtain the object existence state information in the actual visual blind area; wherein the object existence state information includes the relative distance and relative orientation information between the object and the tractor;

[0103] The warning notification module is used to generate corresponding driving notification messages and / or external vehicle alarm reminder messages based on the object existence status information, including:

[0104] Based on the object existence status information, it is determined whether the tractor will collide with the corresponding object; when a collision occurs, a driving notification message including a driving speed reduction and / or an external vehicle alarm reminder message including a preset voice signal is generated.

[0105] The beneficial effect of the above embodiment is that the actual visual blind area is an area that the driver on the tractor cannot see through the naked eye or the rearview mirror. In order to compensate for the invisibility of the actual visual blind area, the actual visual blind area size change information and distribution position change information contained in the actual visual blind area state information are used as a reference to determine the occupied area of ​​the actual visual blind area in the three-dimensional space where the tractor is located (that is, the occupied area formed by the corresponding projection of the actual visual blind area in the three-dimensional space), so as to perform a three-dimensional representation of the actual visual blind area. Based on the occupied area, the radar device on the tractor is instructed to track and detect the occupied area, that is, the radar device is instructed to perform real-time aiming detection on the occupied area to obtain corresponding radar detection data, so as to calculate the existence status information of the object in the actual visual blind area, so as to accurately locate and identify the objects in the area around the tractor. In addition, based on the existence status information of the object, it is determined whether the distance between the tractor and the object is less than a preset distance threshold. If so, it is determined that the tractor will collide with the corresponding object. If not, it is determined that the tractor will not collide with the corresponding object, and when a collision occurs, a driving notification message including a driving speed reduction and / or an external vehicle alarm reminder message including a preset voice signal is generated, so that the driver of the tractor and the personnel located in the actual visual blind spot of the tractor can be reminded and notified in a timely and accurate manner.

[0106] In general, the blind spot warning method and system for heavy-duty semi-trailer tractor determines the factors affecting the formation of the blind spot of the tractor based on the structural information of the heavy-duty semi-trailer tractor, thereby fitting and generating the correlation information between the turning radius and the visual blind spot range of the tractor in different turning scenarios, and continuously marking the changes in the visual blind spot range of the tractor under different turning actions; visually identifying the road ahead of the tractor, predicting the turning action feature information of the tractor in the future driving process, and combining the correlation information to determine the actual visual blind spot azimuth state information of the tractor in the future driving process, accurately predicting the visual blind spot of the tractor, and providing guidance for subsequent radar detection; then based on the actual visual blind spot range state information, instructing the radar equipment on the tractor to perform detection, and obtaining the object existence state information within the actual visual blind spot range, thereby generating a driving notification message and / or an external alarm reminder message, and using the radar detection method as an auxiliary detection method for the visual blind spot of the tractor to achieve global and effective detection of the visual blind spot, thereby improving the driving safety of the tractor.

[0107] The above is only a specific implementation of the present invention, and any other improvements made based on the concept of the present invention are considered to be within the protection scope of the present invention.

Claims

1. A blind spot warning method for a heavy-duty semi-trailer tractor, characterized in that: include: Based on the structural information of the heavy-duty semi-trailer tractor, determining the factors affecting the formation of the blind spot of the tractor; Based on the factors influencing the formation of the blind spot, fitting and generating correlation information between the turning radius of the tractor and the range of the visual blind spot in different turning scenarios; Acquire a road image ahead of the tractor, analyze the road image ahead, and predict characteristic information of turning actions of the tractor in the future driving process; determine actual visual blind spot range state information of the tractor in the future driving process based on the turning action characteristic information and the associated information; Based on the actual visual blind area range status information, instruct the radar device on the tractor to detect the corresponding surrounding area of ​​the tractor, so as to obtain the object existence status information within the actual visual blind area; Based on the object existence status information, generate a corresponding driving notification message and / or an external vehicle alarm reminder message; Based on the structural information of the heavy-duty semi-trailer tractor, determining the factors affecting the formation of the blind spot of the tractor; Based on the factors influencing the formation of the blind spot, the correlation information between the turning radius of the tractor and the range of the visual blind spot in different turning scenarios is generated by fitting, including: Based on the length dimension information of the heavy semi-trailer tractor, the inner wheel difference of the tractor is determined, and based on the shape dimension information of the rearview mirror of the tractor, the field of view range of the rearview mirror of the tractor is determined, so that the inner wheel difference and the field of view range of the rearview mirror are determined as the influencing factors of the blind spot formation of the tractor; Based on the inner wheel difference and the field of view of the rearview mirror, the visual blind area ranges corresponding to the tractor in several turning radius scenarios are determined; wherein the several turning radius scenarios correspond to several discrete turning radii with the same radius difference; neural network learning is performed based on all the visual blind area ranges corresponding to the several turning radius scenarios, and fitting is performed to generate association information between the turning radius of the tractor and the size and distribution position of the visual blind area range in different turning scenarios in which the turning radius continuously changes; Acquiring a road image ahead of the tractor, analyzing the road image ahead, and predicting characteristic information of a turning action of the tractor in the future driving process; and determining actual visual blind area range state information of the tractor in the future driving process based on the turning action characteristic information and the associated information, including: Scan and photograph the road ahead of the tractor to obtain a panoramic image of the road ahead; perform road contour recognition processing on the panoramic image of the road ahead to obtain extended shape information of the road ahead of the tractor; based on the extended shape information of the road and the driving speed of the tractor, predict the change information of the turning radius corresponding to the turning action of the tractor in the future driving process, and use this as the turning action feature information; The turning radius change information is compared and analyzed with the associated information to determine the actual visual blind spot range size change information and distribution position change information corresponding to the turning action of the tractor in the future driving process, and use this as the actual visual blind spot range state information.

2. The heavy-duty semi-trailer tractor blind spot warning method according to claim 1, characterized in that: Based on the actual visual blind area range status information, instruct the radar device on the tractor to detect the corresponding surrounding area of ​​the tractor, so as to obtain the object existence status information within the actual visual blind area; Based on the object existence status information, a corresponding driving notification message and / or an external vehicle alarm reminder message is generated, including: Based on the actual visual blind area size change information and distribution position change information included in the actual visual blind area state information, determine the occupied area range of the actual visual blind area in the three-dimensional space where the tractor is located; based on the occupied area range, instruct the radar equipment on the tractor to track and detect the occupied area range to obtain corresponding radar detection data; then analyze the radar detection data to obtain the object existence state information in the actual visual blind area; wherein the object existence state information includes the relative distance and relative orientation information between the object and the tractor; Based on the object existence status information, it is determined whether the tractor will collide with the corresponding object; when a collision occurs, a driving notification message including a driving speed reduction and / or an external vehicle alarm reminder message including a preset voice signal is generated.

3. The blind spot warning method for a heavy-duty semi-trailer tractor according to claim 2, characterized in that: Based on the occupied area range, instructing the radar device on the tractor to track and detect the occupied area range, including: Step S1, establish a spatial rectangular coordinate system in the occupied area, so that the points inside it can be expressed by coordinates, and use the following formula (1) to determine the scanning range of the radar device on the tractor according to the occupied area: R=1.5×max a,b∈Q {L[(x a ,y a ,z a ),(x b ,y b ,z b )]} (1) In the above formula (1), R indicates that the scanning range of the radar device on the tractor is a spherical area with a radius of R; a,b∈Q indicates that both position a and position b are points within the occupied area; (x a ,y a ,z a ) represents the coordinate point a within the occupied area; (x b ,y b ,z b ) represents the coordinate point b within the occupied area; L[(x a ,y a ,z a ),(x b ,y b ,z b )] indicates the spatial distance between the coordinate points of position a and position b; max a,b∈Q {L[(x a ,y a ,z a ),(x b ,y b ,z b )]} means to obtain the maximum value of the spatial distance between the coordinate points of position a and position b under the condition that a, b∈Q is satisfied; Step S2, using the following formula (2), according to the scanning range of the radar device on the tractor and the single-point scanning range of the radar device, determine the mobile scanning mode of the radar device, In the above formula (2), Indicates that the radar device moves from the current radar device position (x0, y0, z0) to the position point Perform reciprocating movement of L distance for mobile scanning, where L = (R-r0); r0 represents the single-point scanning radius of the radar device on the tractor; Step S3, using the following formula (3), according to the scanning range of the radar device on the tractor and the single-point scanning range of the radar device, determine the moving speed of the radar device during the mobile scanning. In the above formula (3), V represents the moving speed of the radar device during the moving scan; V0 represents the minimum moving speed of the radar device during the moving scan; V M Indicates the maximum moving speed of the radar device when moving and scanning.

4. Heavy-duty semi-trailer tractor blind spot warning system, characterized by: include: A blind spot formation influencing factor determination module, used to determine the blind spot formation influencing factors of the tractor based on the structural information of the heavy-duty semi-trailer tractor; A turning radius-blind spot association identification module, used for fitting and generating association information between the turning radius and the visual blind spot range of the tractor in different turning scenarios based on the influencing factors of the blind spot formation; A turning action feature recognition module is used to obtain a road image ahead of the tractor, analyze the road image ahead, and predict turning action feature information that may occur during the tractor's future driving process; An actual visual blind area determination module, used to determine the actual visual blind area range state information of the tractor during future driving based on the turning action feature information and the associated information; A radar detection control module, for instructing the radar device on the tractor to detect the corresponding surrounding area of ​​the tractor based on the actual visual blind area range status information, so as to obtain the object existence status information within the actual visual blind area range; A warning notification module, used to generate a corresponding driving notification message and / or an external vehicle alarm reminder message based on the object existence status information; The blind spot formation influencing factor determination module is used to determine the blind spot formation influencing factors of the tractor based on the structural information of the heavy-duty semi-trailer tractor, including: Based on the length dimension information of the heavy semi-trailer tractor, the inner wheel difference of the tractor is determined, and based on the shape dimension information of the rearview mirror of the tractor, the field of view range of the rearview mirror of the tractor is determined, so that the inner wheel difference and the field of view range of the rearview mirror are determined as the influencing factors of the blind spot formation of the tractor; The turning radius-blind spot association identification module is used to fit and generate association information between the turning radius and the visual blind spot range of the tractor in different turning scenarios based on the influencing factors of the blind spot formation, including: Based on the inner wheel difference and the field of view of the rearview mirror, the visual blind area ranges corresponding to the tractor in several turning radius scenarios are determined; wherein the several turning radius scenarios correspond to several discrete turning radii with the same radius difference; neural network learning is performed based on all the visual blind area ranges corresponding to the several turning radius scenarios, and fitting is performed to generate association information between the turning radius of the tractor and the size and distribution position of the visual blind area range in different turning scenarios in which the turning radius continuously changes; The turning action feature recognition module is used to obtain the road image ahead of the tractor, analyze the road image ahead, and predict the turning action feature information of the tractor in the future driving process, including: Scan and photograph the road ahead of the tractor to obtain a panoramic image of the road ahead; perform road contour recognition processing on the panoramic image of the road ahead to obtain extended shape information of the road ahead of the tractor; based on the extended shape information of the road and the driving speed of the tractor, predict the change information of the turning radius corresponding to the turning action of the tractor in the future driving process, and use this as the turning action feature information; The actual visual blind area determination module is used to determine the actual visual blind area range state information of the tractor during future driving based on the turning action feature information and the associated information, including: The turning radius change information is compared and analyzed with the associated information to determine the actual visual blind spot range size change information and distribution position change information corresponding to the turning action of the tractor in the future driving process, and use this as the actual visual blind spot range state information.

5. The heavy-duty semi-trailer tractor blind spot warning system according to claim 4, characterized in that: The radar detection control module is used to instruct the radar device on the tractor to detect the corresponding surrounding area of ​​the tractor based on the actual visual blind area range status information, so as to obtain the object existence status information within the actual visual blind area, including: Based on the actual visual blind area size change information and distribution position change information included in the actual visual blind area state information, determine the occupied area range of the actual visual blind area in the three-dimensional space where the tractor is located; based on the occupied area range, instruct the radar equipment on the tractor to track and detect the occupied area range to obtain corresponding radar detection data; then analyze the radar detection data to obtain the object existence state information in the actual visual blind area; wherein the object existence state information includes the relative distance and relative orientation information between the object and the tractor; The early warning notification module is used to generate a corresponding driving notification message and / or an external vehicle alarm reminder message based on the object existence status information, including: Based on the object existence status information, it is determined whether the tractor will collide with the corresponding object; when a collision occurs, a driving notification message including a driving speed reduction and / or an external vehicle alarm reminder message including a preset voice signal is generated.

Images