Obstacle projection processing method and electronic device

Abstract

The present disclosure provides a method and electronic device for processing projection of obstacles. The method may include: obtaining a reference line and road parameters of a driving road to be detected, wherein the reference line includes at least one straight road section and one curved road section, and the road parameters include at least: a turning radius at the curved road, a road width, and a maximum perceived distance of obstacles; based on the road parameters, calculating the maximum deflection angle allowed for the reference line of the driving road; adaptively correcting the reference line using the maximum deflection angle; and performing projection processing of obstacles on the corrected reference line. The present invention solves the technical problem of low accuracy in obstacle projection.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing, and in particular to an obstacle projection processing method and electronic device. Background Art

[0002] Currently, when projecting an obstacle, the vertices of the obstacle are selected as feature points, and then the feature points are sequentially converted to the Frenet coordinate system. Subsequently, the feature points are connected in the new coordinate system to recombine them into a polygon. Since the number of feature points is limited and they can only carry very limited information, there are errors in the projected obstacle, resulting in a technical problem of low accuracy in obstacle projection.

[0003] Currently, no effective solution has been proposed for the above-mentioned technical problem of low accuracy in obstacle projection. Summary of the invention

[0004] The embodiments of the present invention provide a projection processing method of an obstacle and an electronic device, so as to at least solve the technical problem of low accuracy in obstacle projection.

[0005] According to one aspect of an embodiment of the present invention, a method for projection processing of an obstacle is provided, comprising: obtaining a reference line and road parameters of a driving road to be detected, wherein the reference line includes at least one straight road section and one curved road section, and the road parameters include at least: a turning radius at the curved road, a road width, and a maximum perception distance of an obstacle; based on the road parameters, calculating a maximum deflection angle allowed for the reference line of the driving road; adaptively correcting the reference line using the maximum deflection angle; and performing projection processing of the obstacle on the corrected reference line.

[0006] Optionally, the reference line is adaptively corrected using the maximum deflection angle, including: determining whether the actual deflection angle of the reference line is less than or equal to the maximum deflection angle; if the actual deflection angle is greater than the maximum deflection angle, cropping the reference line to obtain a new reference line, wherein the actual deflection angle of the new reference line is less than or equal to the maximum deflection angle.

[0007] Optionally, cropping the reference line to obtain a new reference line includes: cropping along one end of the reference line in sequence according to a predetermined length, and after each cropping, if it is determined that the actual deflection angle of the cropped reference line is greater than the maximum deflection angle, continuing the cropping process until the actual deflection angle of the cropped reference line is less than or equal to the maximum deflection angle, and then terminating the cropping process to obtain a new reference line.

[0008] Optionally, cropping the reference line to obtain a new reference line includes: calculating the difference between the actual deflection angle and the maximum deflection angle, determining a cropping point on the reference line based on the difference, cropping the reference line based on the cropping point, and if it is determined that the actual deflection angle of the cropped reference line is less than or equal to the maximum deflection angle, terminating the cropping process to obtain a new reference line.

[0009] Optionally, obtaining a reference line of a driving road to be detected includes: when a vehicle is driving on the driving road, identifying an overall reference line of the driving road from a map, and determining at least one section of the reference line to be detected from the overall reference line based on predetermined conditions, wherein a lane center line of the driving road is used as the overall reference line, and the overall reference line includes at least one section of a curve with curvature.

[0010] Optionally, based on road parameters, the maximum deflection angle allowed for the reference line of the driving road is calculated, including: determining the center of gravity of the reference line at the curve based on the turning radius and road width of the curve, wherein the center of gravity is the spatial critical point of the reference line; obtaining the spatial boundary line of the reference line based on the center of gravity at the curve; calculating the analytical expression of the spatial boundary line according to a geometric formula, and deriving the coordinates of the intersection of the boundary line according to the analytical expression of the boundary line; using the maximum perception distance as a parameter and the intersection coordinates as a critical value, calculating the maximum deflection angle allowed for the reference line.

[0011] Optionally, the method also includes: combining the intersection coordinates of the dividing lines with the positions of the straight lines and the curves in the reference line to divide the spatial area where the reference line is located to obtain multiple areas; if an obstacle is detected in the spatial area, if the projection result of the obstacle occupies multiple areas at the same time, determining that the projection result of the obstacle is an abnormal projection result.

[0012] Optionally, when the projection result is a polygonal object, if all vertices of the polygonal object are located in different regions, the projection result of the obstacle is an abnormal projection result, and if all vertices of the polygonal object fall in the same region, the projection result is a normal projection result.

[0013] According to another aspect of an embodiment of the present invention, a projection processing device for an obstacle is provided, comprising: an acquisition unit, used to acquire a reference line and road parameters of a driving road to be detected, wherein the reference line includes at least one straight road section and one curved road section, and the road parameters include at least: a turning radius at the curved road, a road width, and a maximum perception distance of an obstacle; a calculation unit, used to calculate a maximum deflection angle allowed for the reference line of the driving road based on the road parameters; a correction unit, used to adaptively correct the reference line using the maximum deflection angle; and a projection unit, used to perform projection processing of the obstacle on the corrected reference line.

[0014] According to another aspect of an embodiment of the present invention, an electronic device is also provided. The electronic device is installed in a vehicle, and the electronic device may include: 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 execute the obstacle projection processing method of the embodiment of the present disclosure.

[0015] According to another aspect of an embodiment of the present invention, a vehicle is provided, including: a projection processing method for an obstacle according to an embodiment of the present disclosure.

[0016] In an embodiment of the present invention, a reference line and road parameters of a driving road to be detected are obtained, wherein the reference line includes at least one straight road section and one curved road section, and the road parameters include at least: a turning radius at the curved road, a road width, and a maximum perceived distance of an obstacle; based on the road parameters, a maximum deflection angle allowed for the reference line of the driving road is calculated; the reference line is adaptively corrected using the maximum deflection angle; and obstacle projection processing is performed on the corrected reference line. That is, in the present application, a reference line that can avoid projection errors at a U-shaped curve can be calculated through geometric modeling using the U-shaped curve turning radius, the maximum obstacle perceived distance, and the maximum road width as parameters. The maximum deflection angle is used to adaptively correct the reference line. The calculation of the maximum deflection angle can be performed before the vehicle travels, and the calculation result can be stored in the map information as a road parameter, thereby effectively improving the passability of vehicles (including autonomous driving vehicles) in U-shaped bends without changing the calculation amount of the original algorithm, thereby ensuring that the feature points of the obstacle are projected on one side of the U-shaped bend, thereby solving the technical problem of low accuracy of obstacle projection and achieving the technical effect of improving the accuracy of obstacle projection.

[0017] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0019] Figure 1 is a flow chart of a method for processing projection of an obstacle according to an embodiment of the present disclosure;

[0020] Figure 2 is a schematic diagram of U-bend scene modeling according to an embodiment of the present disclosure;

[0021] Figure 3 is a schematic diagram of a Frenet coordinate system in the related art;

[0022] Figure 4 is a schematic diagram of projection conditions of obstacles under different reference lines according to an embodiment of the present disclosure;

[0023] Figure 5 is a schematic diagram of a space dividing line according to an embodiment of the present disclosure;

[0024] Figure 6 is a schematic diagram of an analytical formula for calculating arcs according to an embodiment of the present disclosure;

[0025] Figure 7 is a schematic diagram of a spatial region division according to an embodiment of the present disclosure;

[0026] Figure 8 is a schematic diagram of a projection processing device for an obstacle according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0029] Example 1

[0030] In the related art, point-to-point projection is performed only according to the basic definition of Frenet, without considering that the obstacle, as an entity occupying a certain space, may be overstretched or overcompressed after being projected. The projection method in the related art has a small error when facing a low-curvature reference line, but as the curvature of the reference line increases, especially the U-turn that is common in urban roads, the feature points extracted from the obstacle may be projected to the beginning and end of the U-turn reference line, resulting in a serious destruction of the spatial relationship between the main vehicle and the obstacle.

[0031] In this application, in order to solve the projection problem of obstacles at U-turns, the turning radius of the U-turn, the maximum obstacle perception distance, and the maximum road width are used as parameters. The maximum deflection angle of the reference line that can avoid projection errors at the U-turn is calculated through geometric modeling, and the reference line is adaptively corrected using the maximum deflection angle, thereby ensuring that the characteristic points of the obstacle are forced to be projected on one side of the U-turn. The calculation of the maximum deflection angle can be performed before the vehicle is driven, and the calculation result is stored in the map information as a road parameter, thereby effectively improving the passability of autonomous driving vehicles at U-turns without changing the calculation amount of the original algorithm.

[0032] Figure 1 is a flowchart of a method for processing obstacle projection according to an embodiment of the present disclosure. Figure 1 As shown, the method may include the following steps:

[0033] Step S101, obtaining a reference line and road parameters of a driving road to be detected, wherein the reference line includes at least one straight road section and one curved road section, and the road parameters include at least: a turning radius at the curved road, a road width, and a maximum perceived distance of obstacles.

[0034] In the technical solution provided in the above step S101 of the present disclosure, the consistency of the spatial position of the obstacle before and after the projection is used as the judgment criterion, and a suitable reference deflection angle is calculated according to the road parameters, so that the positional relationship between the obstacle represented by the feature point and the road and the vehicle is consistent with the Cartesian coordinate system, providing a data basis for the subsequent lateral and longitudinal decisions of the autonomous driving algorithm.

[0035] In this embodiment, a reference line and road parameters of the driving road to be detected can be obtained, wherein the reference line can include at least one straight section and one curved section. When the reference line is a straight section, the projection result of the Frenet coordinate system is the same as the projection result of the Cartesian coordinate system. When the reference line is a curved section, in order to ensure the accuracy of the obstacle projection, the projection error problem can be solved by limiting the maximum deflection angle of the reference line.

[0036] For example, Figure 2 is a schematic diagram of a U-bend modeling according to an embodiment of the present disclosure, such as Figure 2 As shown, the reference line segments AB and CD are composed of two straight lines and an arc AC. The center of the arc AC is O1 and the radius is r. Assume that the starting point of the reference line is point B and the end point is point D. If the reference line is directly used as the reference line of the Frenet coordinate system, it is easy to cause projection errors. In this application, the projection error problem is solved by limiting the maximum deflection angle of the reference line. Assume that the maximum deflection angle of the reference line with the ball is 180-θ. The part of the reference line that meets the angle requirement is represented by the solid line EAB, and the dotted line ECD that exceeds the angle limit is the cropped part.

[0037] Step S102: Based on the road parameters, a maximum deflection angle allowed for the reference line of the driving road is calculated.

[0038] In the technical solution provided in the above step S102 of the present disclosure, the road parameters may be a U-turn radius, a maximum obstacle perception distance, and a maximum road width.

[0039] In this embodiment, the maximum deflection angle allowed for the reference line of the driving road can be calculated based on the road parameters. For example, the maximum deflection angle of the reference line that can avoid projection errors at the U-turn can be calculated through geometric modeling using the U-turn turning radius, the maximum obstacle perception distance and the maximum road width as parameters.

[0040] In this embodiment, the calculation of the maximum deflection angle can be performed before the vehicle is driven, and the calculation result can be stored in the map information as a road parameter, thereby effectively improving the passability of the vehicle (which may include an autonomous driving vehicle) in the U-turn without changing the calculation amount of the original algorithm.

[0041] Step S103, adaptively correcting the reference line using the maximum deflection angle.

[0042] In the technical solution provided in the above step S103 of the present disclosure, the reference line can be adaptively corrected using the maximum deflection angle. For example, after the maximum deflection angle of the reference line is obtained by calculating the road parameters, the reference line can be adaptively corrected using the maximum deflection angle to ensure that the characteristic points of the obstacle are projected on one side of the U-shaped bend.

[0043] Step S104: performing obstacle projection processing on the corrected reference line.

[0044] In this embodiment, obstacle projection processing can be performed on the corrected reference line. For example, after obtaining the maximum deflection angle of the reference line, when the deflection angle of the actual reference line is less than the maximum deflection angle, the accuracy of the obstacle conversion from the Cartesian coordinate system to the Frenet coordinate system is guaranteed.

[0045] Through the above steps S101 to S104, the reference line and road parameters of the driving road to be detected are obtained, wherein the reference line includes at least one straight road section and one curved road section, and the road parameters include at least: the turning radius at the curved road, the road width, and the maximum perceived distance of obstacles; based on the road parameters, the maximum deflection angle allowed for the reference line of the driving road is calculated; the reference line is adaptively corrected using the maximum deflection angle; and the projection processing of obstacles is performed on the corrected reference line. That is, in this application, the U-turn turning radius, the maximum perceived distance of obstacles, and the maximum road width can be used as parameters. , the maximum deflection angle of the reference line that can avoid projection errors at the U-shaped bend is calculated through geometric modeling, and the reference line is adaptively corrected using the maximum deflection angle. The calculation of the maximum deflection angle can be performed before the vehicle travels, and the calculation result can be stored in the map information as a road parameter, thereby effectively improving the passability of vehicles (including autonomous driving vehicles) at the U-shaped bend without changing the calculation amount of the original algorithm, thereby ensuring that the feature points of the obstacle are projected on one side of the U-shaped bend, thereby solving the technical problem of low accuracy of obstacle projection and achieving the technical effect of improving the accuracy of obstacle projection.

[0046] The above method of this embodiment is further described in detail below.

[0047] As an optional implementation, step S103, using the maximum deflection angle to adaptively correct the reference line, the method includes: determining whether the actual deflection angle of the reference line is less than or equal to the maximum deflection angle; if the actual deflection angle is greater than the maximum deflection angle, cropping the reference line to obtain a new reference line, wherein the actual deflection angle of the new reference line is less than or equal to the maximum deflection angle.

[0048] In this embodiment, it can be determined whether the actual deflection angle of the reference line is less than or equal to the maximum deflection angle. If the actual deflection angle is greater than the maximum deflection angle, the reference line is cropped to obtain a new reference line, wherein the actual deflection angle of the new reference line is less than or equal to the maximum deflection angle. For example, the projection error problem is solved by limiting the maximum deflection angle of the reference line. Assuming that the maximum deflection angle of the reference line is 180-θ, the portion of the reference line that meets the angle requirement is represented by a solid line EAB, and the dotted line ECD that exceeds the angle limit is the cropped portion.

[0049] As an optional implementation, cropping the reference line to obtain a new reference line includes: cropping along one end of the reference line in sequence according to a predetermined length, and after each cropping, if it is determined that the actual deflection angle of the cropped reference line is greater than the maximum deflection angle, continuing the cropping process until the actual deflection angle of the cropped reference line is less than or equal to the maximum deflection angle, and then terminating the cropping process to obtain a new reference line.

[0050] In this embodiment, the reference line can be cut in sequence according to a predetermined length along one end. After each cutting, if it is determined that the actual deflection angle of the reference line after cutting is greater than the maximum deflection angle, the cutting process is continued until the actual deflection angle of the reference line after cutting is less than or equal to the maximum deflection angle, and the cutting process is terminated to obtain a new reference line. For example, after determining that the actual deflection angle is greater than the maximum deflection angle, the length of the reference line to be cut can be pre-set, and then, the reference line can be cut in sequence according to the predetermined length along one end until the actual deflection angle of the reference line after cutting is less than or equal to the maximum deflection angle.

[0051] As an optional implementation, cropping the reference line to obtain a new reference line includes: calculating the difference between the actual deflection angle and the maximum deflection angle, determining a cropping point on the reference line based on the difference, cropping the reference line based on the cropping point, and if it is determined that the actual deflection angle of the cropped reference line is less than or equal to the maximum deflection angle, terminating the cropping process to obtain a new reference line.

[0052] In this embodiment, the difference between the actual deflection angle and the maximum deflection angle can be calculated, and a clipping point can be determined on the reference line based on the difference. The reference line is clipped based on the clipping point. If it is determined that the actual deflection angle of the clipped reference line is less than or equal to the maximum deflection angle, the clipping process is terminated to obtain a new reference line. For example, the difference between the actual deflection angle and the maximum deflection angle is calculated. When it is determined that the actual deflection angle of the clipped reference line is greater than the maximum deflection angle, the clipping point on the reference line is determined, and clipping is performed from the tail of the reference line, that is, the part of the reference line that exceeds the angle limit is clipped, until it is determined that the actual deflection angle of the clipped reference line is less than or equal to the maximum deflection angle, then the clipping process is terminated to obtain a new reference line.

[0053] As an optional implementation, step S101, obtaining a reference line of a driving road to be detected, includes: when a vehicle is driving on the driving road, identifying an overall reference line of the driving road from a map, and determining at least one section of the reference line to be detected from the overall reference line based on predetermined conditions, wherein the lane center line of the driving road is used as the overall reference line, and the overall reference line includes at least one section of a curve with curvature.

[0054] In this embodiment, when the vehicle is traveling on the driving road, the overall reference line of the driving road is identified from the map, and at least one section of the reference line to be detected is determined from the overall reference line based on predetermined conditions. For example, when the vehicle is traveling on a road including at least one curve, the center line of the lane of the driving road is identified from the map, the center line of the lane is used as the overall reference line, and at least one section of the reference line to be detected is determined from the overall reference line based on predetermined conditions.

[0055] In this embodiment, the predetermined condition may be a maximum deflection angle of the reference line, and the maximum deflection angle may be calculated by geometric modeling.

[0056] In this embodiment, the maximum deflection angle of the reference line of the driving road can be calculated before the vehicle travels, and the calculation result can be stored in the map information as a road parameter, thereby effectively improving the vehicle's passability on roads with a certain curvature without changing the calculation amount of the original algorithm.

[0057] As an optional implementation, based on road parameters, the maximum deflection angle allowed for a reference line of the driving road is calculated, and the method includes: determining the center of gravity of the reference line at the curve based on the turning radius and road width of the curve, wherein the center of gravity is the spatial critical point of the reference line; obtaining the spatial boundary line of the reference line based on the center of gravity at the curve; calculating the analytical expression of the spatial boundary line according to a geometric formula, and deriving the coordinates of the intersection of the boundary line according to the analytical expression of the boundary line; using the maximum perception distance as a parameter and the intersection coordinates as a critical value, calculating the maximum deflection angle allowed for the reference line.

[0058] In this embodiment, the center of gravity of the curve in the reference line can be determined based on the turning radius and the road width of the curve in the reference line, and then the spatial boundary line of the reference line can be obtained based on the center of gravity of the curve. The analytical expression of the spatial boundary line is calculated according to the geometric formula, and the intersection coordinates of the boundary line are obtained according to the analytical expression of the boundary line. The maximum perception distance is used as a parameter and the intersection coordinates are used as a critical value to calculate the maximum deflection angle allowed for the reference line. For example, mathematical modeling is performed on the U-bend reference line to establish a spatial boundary line, which divides the entire space into 4 areas. When obstacles occupy multiple areas at the same time, projection errors will be triggered. The analytical expression of the spatial boundary line is calculated according to the geometric formula, and the road width and the turning radius are used as parameters. Combined with the analytical expression of the spatial boundary line, the maximum deflection angle of the reference line is calculated.

[0059] As an optional implementation, the method also includes: combining the intersection coordinates of the dividing lines with the positions of the straight lines and the curves in the reference line to divide the spatial area where the reference line is located to obtain multiple areas; if an obstacle is detected in the spatial area, if the projection result of the obstacle occupies multiple areas at the same time, determining that the projection result of the obstacle is an abnormal projection result.

[0060] In this embodiment, the intersection coordinates of the dividing line can be combined with the position of the straight road and the position of the curve in the reference line to divide the spatial area where the reference line is located into multiple areas. If an obstacle is detected in the spatial area, if the projection result of the obstacle occupies multiple areas at the same time, the projection result of the obstacle is determined to be an abnormal projection result. For example, mathematical modeling is performed on the U-shaped bend reference line to establish a spatial dividing line. The dividing line divides the entire space into 4 areas. When the obstacle occupies multiple areas at the same time, a projection error will be triggered.

[0061] As an optional implementation, when the projection result is a polygonal object, if all vertices of the polygonal object are located in different areas, the projection result of the obstacle is an abnormal projection result; if all vertices of the polygonal object fall in the same area, the projection result is a normal projection result.

[0062] In this embodiment, when the projection result is a polygonal object, if all vertices of the polygonal object are located in different areas, the projection result of the obstacle is an abnormal projection result, and if all vertices of the polygonal object fall in the same area, the projection result is a normal projection result. For example, when the obstacle exists as a polygon, if all vertices of the polygon fall in the same area, there will be no projection error. If there are vertices located in different areas, it will cause a projection error.

[0063] In the above embodiment of the present disclosure, it is determined whether the actual deflection angle of the reference line is less than or equal to the maximum deflection angle; if the actual deflection angle is greater than the maximum deflection angle, the reference line is clipped to obtain a new reference line, wherein the actual deflection angle of the new reference line is less than or equal to the maximum deflection angle; when the vehicle is traveling on the driving road, the overall reference line of the driving road is identified from the map, and at least one section of the reference line to be detected is determined from the overall reference line based on predetermined conditions, wherein the lane centerline of the driving road is used as the overall reference line, and the overall reference line includes at least one section of a curve with curvature; based on the turning radius and the road width at the curve in the reference line, the center of gravity at the curve in the reference line is determined, wherein the center of gravity is the spatial critical point of the reference line; based on the center of gravity at the curve, the spatial boundary line of the reference line is obtained; the analytical expression of the spatial boundary line is calculated according to the geometric formula, and the boundary line analytical expression is used to obtain the boundary line. The coordinates of the intersection of the boundary lines; using the maximum perception distance as a parameter and the intersection coordinates as a critical value, the maximum deflection angle of the reference line allowed to be deflected is calculated. That is to say, in the present application, the U-turn radius, the maximum obstacle perception distance and the maximum road width can be used as parameters, and the maximum deflection angle of the reference line that can avoid projection errors at the U-turn can be calculated through geometric modeling. The reference line is adaptively corrected using the maximum deflection angle. The calculation of the maximum deflection angle can be performed before the vehicle travels, and the calculation result can be stored in the map information as a road parameter, thereby effectively improving the passability of vehicles (including autonomous driving vehicles) in the U-turn without changing the calculation amount of the original algorithm, thereby ensuring that the characteristic points of the obstacle are projected on one side of the U-turn, thereby solving the technical problem of low accuracy of obstacle projection and achieving the technical effect of improving the accuracy of obstacle projection.

[0064] Example 2

[0065] The obstacle projection processing method disclosed in the present invention is further introduced below in conjunction with preferred embodiments.

[0066] The safe driving of self-driving cars depends on accurate perception information and efficient planning algorithms. When self-driving cars are driving on regular roads, the position information of the vehicle itself and surrounding obstacles will be converted from the Cartesian coordinate system to the Frenet coordinate system. The Frenet coordinate system uses the center line of the lane of the urban road as the reference line and projects the points in the Cartesian coordinate system onto the reference line. The obstacles around the vehicle exist in the form of point clouds after being perceived by the radar. If each point in the point cloud is projected, the calculation time of the algorithm will be greatly increased; if the point cloud is simplified to a few feature points, the obstacle contour information after projection will have errors, especially in the U-bend scene, the obstacle boundary after projection may be stretched several times, so it is necessary to design an efficient and low-error Frenet projection method for obstacles at the U-bend.

[0067] In the related art, point-to-point projection is performed according to the basic definition of Frenet, without considering that the obstacle may be overstretched or overcompressed after being projected as an entity occupying a certain space. The projection method in the related art has a small error when facing a low-curvature reference line, but as the curvature of the reference line increases, especially the U-turn that is common in urban roads, the feature points extracted from the obstacle may be projected to the beginning and end of the U-turn reference line, resulting in a serious destruction of the spatial relationship between the main vehicle and the obstacle.

[0068] This application aims at the projection problem of obstacles at U-shaped bends and proposes a U-shaped bend obstacle projection method based on adaptive correction of reference lines. This method uses the turning radius of the U-shaped bend and the maximum road width as parameters, calculates the maximum deflection angle of the reference line that can avoid projection errors at the U-shaped bend through geometric modeling, and uses the maximum deflection angle to adaptively correct the reference line, thereby ensuring that the characteristic points of the obstacle are forced to be projected on one side of the U-shaped bend. This method is easy to implement, and the calculation of the maximum deflection angle can be performed before the vehicle is driven, and the calculation result is stored in the map information as a road parameter, thereby effectively improving the passability of autonomous driving vehicles in U-shaped bends without changing the calculation amount of the original algorithm.

[0069] Cartesian coordinate systems are usually used to describe the spatial position of objects. However, in autonomous driving, Cartesian coordinate systems can only give the position coordinates of the vehicle, and cannot reflect the deviation distance between the vehicle and the center line of the lane and the distance the vehicle has traveled on the road. In reality, most urban roads are arcs with a certain curvature. When a vehicle is driving on the road, it only needs to pay attention to the position relationship between the vehicle and the surrounding lanes. Based on the above factors, the Frenet coordinate system is usually used instead of the Cartesian coordinate system.

[0070] Figure 3 is a schematic diagram of a Frenet coordinate system in the related art, such as Figure 3As shown in Figure 1, the Frenet coordinate system describes the spatial relationship between the vehicle and the road. The coordinate system uses the lane centerline as the reference line, which can be an arbitrary curve. The coordinate axis is generated based on the tangent vector (longitudinal) and the lateral vector (lateral) of the reference line. The Frenet coordinate is expressed as (s, d), where s represents the distance the vehicle travels on the coordinate axis, and d represents the distance the vehicle deviates from the reference line laterally. In the Frenet coordinate system, the trajectory of the vehicle in the lane can be simplified to a straight line, which reduces the amount of data processing calculations for subsequent longitudinal and lateral decisions. The point-to-point conversion between the Cartesian coordinate system and the Frenet coordinate system is accurate and reliable, but obstacles are polygons composed of countless points. In order to ensure the computational efficiency of the algorithm, the obstacles must be simplified into several feature points. Existing methods take the vertices of the obstacle as feature points, sequentially convert the feature points to the Frenet coordinate system, and then connect the feature points in the new coordinate system to reassemble them into a polygon. The number of feature points is limited and can only carry very limited information, so there must be errors in the projected obstacles.

[0071] Figure 4 is a schematic diagram of projection conditions of obstacles under different reference lines according to an embodiment of the present disclosure, such as Figure 4 As shown in the figure, the projection of the same obstacle on reference lines with different curvatures. The obstacle is simplified into a line segment, and the feature points are the two endpoints A and B of the line segment. In the Cartesian coordinate system, the obstacle is located behind the vehicle and does not intersect with the vehicle. The obstacle is projected onto three reference lines with different curvatures. According to the projection results, it can be found that when the reference line is a standard straight line, the projection result is the same as that of the Cartesian coordinate system. When the reference line has a certain curvature, the distance traveled by the vehicle projected at point B on the coordinate axis will increase, causing the line AB to pass through the vehicle or be located in front of the vehicle. The greater the curvature of the reference line, the greater the obstacle projection deviation. This deviation cannot be corrected by changing the point-to-point projection algorithm, but can be corrected by limiting the deflection angle of the reference line.

[0072] In this application, the consistency of the spatial relationship of the obstacle before and after projection is used as the judgment criterion, and the appropriate reference line deflection angle is calculated according to the road parameters, so that the position relationship between the obstacle represented by the feature point and the road and vehicle is consistent with the Cartesian coordinate system, providing a data basis for the subsequent lateral and longitudinal decisions of the autonomous driving algorithm.

[0073] In order to ensure the accuracy and reliability of obstacle projection, the present invention provides a U-bend obstacle projection method based on adaptive correction of reference lines.

[0074] The U-bend modeling process of this method is introduced below.

[0075] like Figure 2As shown in the figure, the bold line is the reference line of the U-shaped bend, and the reference line is composed of two straight lines, line segment AB and line segment CD, and arc AC. The center of arc AC is O1, the radius is r, and the starting point of the reference line is set as point B and the end point is set as point D. Directly using the reference line as the reference line of the Frenet coordinate system is prone to cause projection errors. The present invention solves this problem by limiting the maximum deflection angle of the reference line. The maximum deflection angle of the reference line to be determined is set to 180-θ. The part of the reference line that meets the angle requirement is represented by the solid line EAB, and the dotted line ECD that exceeds the angle limit is the clipped part.

[0076] The process of creating spatial boundaries using this method is introduced below.

[0077] Figure 5 is a schematic diagram of a space dividing line according to an embodiment of the present disclosure, such as Figure 5 As shown, first remove Figure 3 The dotted part of the image and find the critical point O in space i , where i = 1, 2, ... n, point O i The position relationship needs to meet the following conditions: i Draw a perpendicular line to line segment AB, with point G as its foot. i is the center of the circle, line segment O i Draw a circle with the length R of G as the radius. The circle must intersect with point E. All points O that meet the requirements i An arc can be formed. To limit the length of the arc, the present invention takes the center O1 as the starting point of the arc and the point O on the extension line of FE as the starting point of the arc. n As the end point, through the above operation, two space dividing lines are obtained, which are line segment O n E and arc O1O n .

[0078] The following is the calculation line segment O of this method n The analytical process of E is introduced.

[0079] A Cartesian coordinate system is established with point A as the origin. According to geometric calculation, the coordinates of point E are (a, b). The calculation formula is as follows:

[0080] a=-rsinθ

[0081] b=r+rcosθ

[0082] Among them, a can be used to represent the horizontal coordinate of point E, b can be the vertical coordinate of point E, and r is Figure 5 The center of the circle is O i The radius of the arc.

[0083] According to geometric calculation, the abscissa of point F can be obtained, which is expressed as follows:

[0084]

[0085] The ordinate of point F is 0. Based on points E and F, we get line segment O. n The analytical expression of E is as follows:

[0086]

[0087] The following is the calculation arc O1O of this method n The analytical process is introduced in this paper.

[0088] Figure 6 is a schematic diagram of a calculation arc analytical formula according to an embodiment of the present disclosure, such as Figure 6 As shown, follow the calculation line segment O n The Cartesian coordinate system established in the process of solving the analytical expression of E, delete the redundant auxiliary lines, connect the line segment GE, and pass through O i Draw a perpendicular line from point E to line segment GE, with the foot of the perpendicular being H. Point H is also the midpoint of line segment GE. Draw a perpendicular line from point E to the horizontal axis and intersect it at point I. Set point O i The coordinates are (x′, y′). Since the dotted triangle IGE and the solid triangle HGO i are similar triangles and can be expressed as follows:

[0089]

[0090] Calculate arc O1O n The analytical formula is as follows:

[0091]

[0092] Among them, a can be the horizontal coordinate of point E, b can be the vertical coordinate of point E, and x′ can be the vertical coordinate of point O. i The horizontal axis of .

[0093] The execution space region division process of this method is introduced below.

[0094] Figure 7 is a schematic diagram of a spatial region division according to an embodiment of the present disclosure, such as Figure 7 As shown in FIG. 1 , the spatial boundary obtained by the above method divides the entire space into four areas, namely area 1, 2, 3, and other areas. Assuming that the vehicle moves from point B to point E, area 1 represents line segment DO n On the left side, area 2 represents line segment DO n Area 3 represents the right side of line segment BA. When the obstacle exists as a polygon, if all the vertices of the polygon fall in the same area, there will be no projection error. If there are vertices located in different areas, it will definitely lead to projection errors.

[0095] The following is an introduction to the process of calculating the angle θ based on the input parameters of this method.

[0096] The obstacle is located at O n When nearby, it is easy to trigger the projection error. Just use O n The position of is beyond the road range. Assuming that the maximum width of the road is L meters, to reduce the complexity of the algorithm, n The analytical expression of E and the arc O1O n The analytical expressions of are differentiated separately to obtain the horizontal coordinates when the derivatives of the two functions are the same, and the horizontal coordinates are substituted into the line segment O n E, and assuming that the calculated result is equal to L meters, the value of angle θ can be solved. The functional relationship expression of angle θ can be as follows:

[0097]

[0098] Among them, L is the maximum width of the road, r is Figure 5 The center of the circle is O i is the radius of the arc, and θ is the complementary angle of the maximum deflection angle.

[0099] After obtaining the angle θ, we can Figure 3 As shown, the reference line is adaptively corrected, and the reference line is clipped from the tail to make the deflection angle of the reference line less than 180-θ, and it is used as the reference line of the U-bend and obstacle projection is performed.

[0100] In this embodiment, mathematical modeling is performed on the U-bend reference line; a spatial dividing line is established, which divides the entire space into 4 areas. When obstacles occupy multiple areas at the same time, projection errors will be triggered; the analytical expression of the spatial dividing line is calculated according to the geometric formula; the maximum deflection angle of the reference line is calculated by combining the analytical expression of the spatial dividing line with the road width and the turning radius as parameters; the reference line is adaptively corrected using the maximum deflection angle of the reference line, and Frenet projection is performed on the corrected reference line. The obstacle spatial relationship under the Frenet coordinate system obtained by the present invention is consistent with the Cartesian coordinate system, which ensures the accuracy and efficiency of the autonomous driving algorithm, thereby solving the technical problem of low accuracy of obstacle projection and achieving the technical effect of improving the accuracy of obstacle projection.

[0101] Example 3

[0102] The present disclosure also provides a method for executing Figure 1 The obstacle projection method of the illustrated embodiment is an obstacle projection processing device.

[0103] Figure 8 is a schematic diagram of an obstacle projection processing device according to an embodiment of the present disclosure, such as Figure 8 As shown, the obstacle projection processing device 800 may include: an acquisition unit 801 , a calculation unit 802 , a correction unit 803 and a projection unit 804 .

[0104] The acquisition unit 801 is used to acquire a reference line and road parameters of a driving road to be detected, wherein the reference line includes at least one straight road section and one curved road section, and the road parameters include at least: a turning radius at the curved road, a road width, and a maximum perceived distance of an obstacle;

[0105] A calculation unit 802 is used to calculate a maximum deflection angle allowed for a reference line of the driving road based on the road parameters;

[0106] A correction unit 803, configured to adaptively correct the reference line using the maximum deflection angle;

[0107] The projection unit 804 is used to perform obstacle projection processing on the corrected reference line.

[0108] Optionally, the correction unit 803 includes: a judgment module and a clipping module. The judgment module is used to judge whether the actual deflection angle of the reference line is less than or equal to the maximum deflection angle; the clipping module is used to clip the reference line if the actual deflection angle is greater than the maximum deflection angle to obtain a new reference line, wherein the actual deflection angle of the new reference line is less than or equal to the maximum deflection angle.

[0109] The cutting module includes a first cutting submodule and a second cutting submodule. The first cutting submodule is used to cut in sequence along one end of the reference line according to a predetermined length. After each cutting, if it is determined that the actual deflection angle of the reference line after cutting is greater than the maximum deflection angle, the cutting process is continued until the actual deflection angle of the reference line after cutting is less than or equal to the maximum deflection angle, the cutting process is terminated, and a new reference line is obtained; the second cutting submodule is used to calculate the difference between the actual deflection angle and the maximum deflection angle, determine the cutting point on the reference line based on the difference, cut the reference line based on the cutting point, and if it is determined that the actual deflection angle of the reference line after cutting is less than or equal to the maximum deflection angle, the cutting process is terminated to obtain a new reference line.

[0110] Optionally, the acquisition unit 801 includes a first determination module. The first determination module is used to identify the overall reference line of the driving road from the map when the vehicle is driving on the driving road, and determine at least one section of the reference line to be detected from the overall reference line based on a predetermined condition, wherein the lane center line of the driving road is used as the overall reference line, and the overall reference line includes at least one section of a curve with curvature.

[0111] Optionally, the calculation unit 802 includes: a second determination module, an acquisition module, a first calculation module, and a second calculation module. The second determination module is used to determine the center of gravity at the bend in the reference line based on the turning radius and the road width at the bend in the reference line, wherein the center of gravity is the spatial critical point of the reference line; the acquisition module is used to obtain the spatial boundary line of the reference line based on the center of gravity at the bend; the first calculation module is used to calculate the analytical expression of the spatial boundary line according to the geometric formula, and obtain the intersection coordinates of the boundary line according to the analytical expression of the boundary line; the second calculation module is used to calculate the maximum deflection angle allowed for the reference line with the maximum perception distance as a parameter and the intersection coordinates as a critical value.

[0112] Optionally, the device further includes: a division module and a third determination module. The division module is used to divide the spatial area where the reference line is located into multiple areas by combining the intersection coordinates of the boundary lines with the positions of the straight road and the curve in the reference line; the third determination module is used to determine that the projection result of the obstacle is an abnormal projection result if an obstacle is detected in the spatial area and the projection result of the obstacle occupies multiple areas at the same time.

[0113] Among them, when the projection result is a polygonal object, if all vertices of the polygonal object are located in different areas, the projection result of the obstacle is an abnormal projection result, and if all vertices of the polygonal object fall in the same area, the projection result is a normal projection result.

[0114] In the obstacle projection processing device of the above-mentioned embodiment of the present disclosure, a reference line and road parameters of a driving road to be detected are acquired through an acquisition unit, wherein the reference line includes at least one straight road section and one curved road section, and the road parameters include at least: a turning radius at the curve, a road width, and a maximum perception distance of an obstacle; a calculation unit calculates a maximum deflection angle allowed for the reference line of the driving road based on the road parameters; a correction unit adaptively corrects the reference line using the maximum deflection angle; and a projection unit performs obstacle projection processing on the corrected reference line, and adaptively expands and contracts the head and tail parts of the reference line, thereby solving the technical problem of low accuracy of obstacle projection and achieving the technical effect of improving the accuracy of obstacle projection.

[0115] Example 4

[0116] According to another aspect of an embodiment of the present invention, an electronic device is also provided. The electronic device is installed in a vehicle, and the electronic device may include: 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 execute the obstacle projection processing method of the embodiment of the present disclosure.

[0117] Example 5

[0118] According to an embodiment of the present disclosure, the present disclosure also provides a vehicle, including the obstacle projection processing method of the embodiment of the present disclosure.

[0119] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0120] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0121] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, units or models, which can be electrical or other forms.

[0122] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0123] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0124] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the relevant technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0125] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for processing obstacle projection, characterized in that: include: Acquire a reference line and road parameters of a driving road to be detected, wherein the reference line includes at least one straight road section and one curved road section, and the road parameters include at least: a turning radius at the curved road, a road width, and a maximum perceived distance of an obstacle; Based on the road parameters, a maximum deflection angle allowed for the reference line of the driving road is calculated; The reference line is cropped using the maximum deflection angle, wherein the maximum deflection angle is used to indicate a maximum angle at which the reference line is allowed to deflect; Perform obstacle projection on the clipped reference line.

2. The method according to claim 1, characterized in that The reference line is clipped using the maximum deflection angle, comprising: Determining whether the actual deflection angle of the reference line is less than or equal to the maximum deflection angle; If the actual deflection angle is greater than the maximum deflection angle, the reference line is clipped to obtain a new reference line, wherein the actual deflection angle of the new reference line is less than or equal to the maximum deflection angle.

3. The method according to claim 2, characterized in that The reference line is trimmed to obtain a new reference line, including: trimming along one end of the reference line in sequence according to a predetermined length, and after each trimming, if it is determined that the actual deflection angle of the reference line after trimming is greater than the maximum deflection angle, the trimming process is continued until the actual deflection angle of the reference line after trimming is less than or equal to the maximum deflection angle, and the trimming process is terminated to obtain the new reference line.

4. The method according to claim 2, characterized in that: Cropping the reference line to obtain a new reference line includes: calculating the difference between the actual deflection angle and the maximum deflection angle, determining a cropping point on the reference line based on the difference, and cropping the reference line based on the cropping point; if it is determined that the actual deflection angle of the reference line after cropping is less than or equal to the maximum deflection angle, terminating the cropping process to obtain the new reference line.

5. The method according to claim 1, characterized in that Obtaining a reference line of a driving road to be detected, comprising: when a vehicle is driving on the driving road, identifying an overall reference line of the driving road from a map, and determining at least one section of the reference line to be detected from the overall reference line based on predetermined conditions, wherein a lane center line of the driving road is used as the overall reference line, and the overall reference line includes at least one section of a curve with curvature.

6. The method according to any one of claims 1 to 5, characterized in that Based on the road parameters, Calculating the maximum deflection angle allowed for the reference line of the driving road includes: Determine the center of gravity of the curve in the reference line based on the turning radius and the road width of the curve in the reference line, wherein the center of gravity is the spatial critical point of the reference line; Based on the center of gravity at the curve, obtaining a spatial boundary line of the reference line; Calculating the analytical expression of the space dividing line according to the geometric formula, and obtaining the coordinates of the intersection point of the dividing line according to the analytical expression of the dividing line; The maximum perception distance is used as a parameter and the intersection coordinates are used as a critical value to calculate the maximum deflection angle of the reference line allowed to be deflected.

7. The method according to claim 6, characterized in that The method further comprises: The intersection coordinates of the dividing lines are combined with the positions of the straight road and the curve in the reference line to divide the space area where the reference line is located into multiple areas; If it is detected that the obstacle appears in the spatial area, and if the projection result of the obstacle occupies the multiple areas at the same time, it is determined that the projection result of the obstacle is an abnormal projection result.

8. The method according to claim 7, characterized in that When the projection result is a polygonal object, if all vertices of the polygonal object are located in different areas, the projection result of the obstacle is an abnormal projection result; if all vertices of the polygonal object fall within the same area, the projection result is a normal projection result.

9. An electronic device, characterized in that: The electronic device is installed in a vehicle, and 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 execute the obstacle projection processing method described in any one of claims 1 to 8.

10. A vehicle, characterized in that: A stored program is run in the vehicle, wherein when the program is run, a processor of the device is controlled to execute the obstacle projection processing method according to any one of claims 1 to 8.

Images