Conical barrel obstacle avoidance method and device, electronic equipment and target vehicle

Abstract

The invention relates to the technical field of intelligent driving, in particular to a cone barrel obstacle avoidance method and device, electronic equipment and a target vehicle. Acquiring a front original image corresponding to the target vehicle; identifying the front original image, determining original lane line information in the front original image, and performing parallel processing on original lane lines to generate parallel lane line information; acquiring position information of the target cone barrel in a vehicle coordinate system; based on the original lane line information and the parallel lane line information, detecting whether the target cone bucket is in a target lane; and controlling the target vehicle according to the detection result. Therefore, the control accuracy of the target vehicle is ensured. Therefore, the problems of mistaken braking and lane changing rollback caused by the fact that the electronic equipment in the target vehicle misjudges that the cone barrel is in the lane or the adjacent lane are avoided. According to the method, the accuracy of intelligent driving is improved, and then the problems that the user is insufficient in confidence of intelligent driving, the utilization rate is reduced and popularization cannot be achieved are solved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving technology, and in particular to a cone-barrel obstacle avoidance method, device, electronic equipment and a target vehicle. Background Art

[0002] With the rapid development of electrification and intelligence, automatic assisted driving technology has become one of the selling points of various car manufacturers. The purpose of assisted driving is not only to minimize the danger of car driving, but also to reduce the heavy driving tasks of users.

[0003] However, when the automatic assisted driving cruise is turned on, the vehicle system has a relatively simple judgment on the positional relationship between the front cone and the lane line, and may misjudge that the cone is in the current lane or the adjacent lane, resulting in problems such as misbraking and lane change backing, which directly leads to users' lack of confidence in smart driving, reduced usage rate, and failure to popularize it. Summary of the invention

[0004] In view of this, the present invention provides a cone barrel obstacle avoidance method, device, electronic device and target vehicle to solve the current problem of misjudging the cone barrel in the own lane or the adjacent lane, resulting in wrong braking and lane change back-up, which directly leads to users' lack of confidence in smart driving, reduced usage rate and non-popularization.

[0005] In a first aspect, the present invention provides a cone barrel obstacle avoidance method, the method comprising:

[0006] Obtain the original front image corresponding to the target vehicle;

[0007] Identify the original image ahead, determine the original lane line information in the original image ahead, and perform parallel processing on the original lane line to generate parallel lane line information;

[0008] Get the position information of the target cone in the vehicle coordinate system;

[0009] Based on the original lane line information and the parallel lane line information, detect whether the target cone is in the target lane;

[0010] Based on the detection results, the target vehicle is controlled.

[0011] The cone barrel obstacle avoidance method provided in the embodiment of the present application obtains the original image in front of the target vehicle; identifies the original image in front, determines the original lane line information in the original image in front, and ensures the accuracy of the original lane line information in the determined original image in front. Then, the original lane line is processed in parallel to generate parallel lane line information, and the accuracy of the generated parallel lane line information is ensured. The position information of the target cone barrel in the vehicle coordinate system is obtained; based on the original lane line information and the parallel lane line information, it is detected whether the target cone barrel is in the target lane, and the accuracy of the obtained detection result is ensured, and it is avoided that only based on the original lane line information to detect whether the target cone barrel is in the target lane, resulting in the accuracy of the obtained detection result. According to the detection result, the target vehicle is controlled, thereby ensuring the accuracy of the control of the target vehicle. This avoids the problem that the electronic equipment in the target vehicle misjudges that the cone barrel is in the lane or the adjacent lane, resulting in the problem of misbraking and lane change back. Therefore, the above method improves the accuracy of intelligent driving, thereby avoiding the problem that users lack confidence in intelligent driving, the usage rate is reduced, and it cannot be popularized.

[0012] In an optional implementation, the original lane line information includes an original fitting function corresponding to each original lane line, and the front original image is identified to determine the original lane line information in the front original image, including:

[0013] The original image in front is input into the lane line detection model, the original lane lines in the original image in front are identified, and the position information and direction information corresponding to each original lane line are determined.

[0014] For each original lane line, output multiple lane line points corresponding to the original lane line according to the position and direction information corresponding to the original lane line;

[0015] Based on the position information of each lane line point corresponding to the original lane line, the original fitting function corresponding to each lane line is fitted and generated.

[0016] The cone obstacle avoidance method provided in the embodiment of the present application inputs the original image in front into the lane line detection model, identifies the original lane lines in the original image in front, determines the position information and direction information corresponding to each original lane line, and ensures the accuracy of the position information and direction information corresponding to each original lane line. For each original lane line, according to the position and direction information corresponding to the original lane line, multiple lane line points corresponding to the original lane line are output, and the accuracy of the multiple lane line points corresponding to the output original lane line is ensured. Based on the position information of each lane line point corresponding to the original lane line, the original fitting function corresponding to each lane line is fitted and generated, and the accuracy of the original fitting function corresponding to each lane line generated by fitting is ensured.

[0017] In an optional implementation, the parallel lane line information includes parallel fitting functions corresponding to each parallel lane line, and the original lane lines are processed in parallel to generate the parallel lane line information, including:

[0018] For any two adjacent original lane lines, determine the median lines corresponding to the two adjacent original lane lines along the vehicle driving direction according to the position information of the two adjacent original lane lines;

[0019] Determine the position information corresponding to multiple median points in the median line;

[0020] According to the position information corresponding to each median point, a median line fitting function is generated by fitting;

[0021] According to the lane width, the median line is offset to both sides to determine the parallel lane lines corresponding to the two adjacent original lane lines;

[0022] Based on each parallel lane line, a parallel fitting function corresponding to each parallel lane line is determined.

[0023] The cone barrel obstacle avoidance method provided in the embodiment of the present application determines the median line along the vehicle's driving direction corresponding to the two adjacent original lane lines based on the position information of the two adjacent original lane lines, thereby ensuring the accuracy of the median line along the vehicle's driving direction corresponding to the determined two adjacent original lane lines. Determine the position information corresponding to multiple median points in the median line; and generate a median fitting function based on the position information corresponding to each median point, thereby ensuring the accuracy of the median fitting function generated by fitting. According to the lane width, the median line is offset to both sides to determine the parallel lane lines corresponding to the two adjacent original lane lines, thereby ensuring the accuracy of the parallel lane lines corresponding to the determined two adjacent original lane lines. Based on each parallel lane line, determine the parallel fitting function corresponding to each parallel lane line, thereby ensuring the accuracy of the parallel fitting function corresponding to the determined parallel lane lines.

[0024] In an optional implementation, the target lane includes the vehicle lane, the left lane of the vehicle, and the right lane of the vehicle. Based on the original lane line information and the parallel lane line information, detecting whether the target cone is in the target lane includes:

[0025] For each target lane, based on the original lane line information, detect whether the target cone is in the target lane to obtain a first detection result;

[0026] According to the parallel lane line information, detect whether the target cone is in the target lane to obtain a second detection result;

[0027] According to the first detection result and the second detection result, it is determined whether the target cone is in the target lane, and the target lane is any one of the lane, the left lane and the right lane corresponding to the target vehicle.

[0028] The cone obstacle avoidance method provided in the embodiment of the present application detects whether the target cone is in the target lane according to the original lane line information for each target lane, and obtains a first detection result, thereby ensuring the accuracy of the first detection result. According to the parallel lane line information, the target cone is detected to be in the target lane, and a second detection result is obtained, thereby ensuring the accuracy of the second detection result. According to the first detection result and the second detection result, it is determined whether the target cone is in the target lane, thereby ensuring the accuracy of the determination of whether the target cone is in the target lane.

[0029] In an optional implementation, the original lane line includes an original left lane line, an original left-left lane line, an original right lane line, and an original right-right lane line, and the original lane line information includes an original left lane line fitting function, an original left-left lane line fitting function, an original right lane line fitting function, and an original right-right lane line fitting function. According to the original lane line information, whether the target cone is in the target lane is detected to obtain a first detection result, including:

[0030] According to the position information of the target cone barrel in the vehicle coordinate system, the target ordinate and the target abscissa corresponding to the target cone barrel are determined;

[0031] If the target ordinate exceeds the ordinate value range of the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function, and the original right-right lane line fitting function, then the output is that it is currently impossible to determine whether the target cone is in the target lane;

[0032] If the target ordinate is within the ordinate value range of the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function and the original right-right lane line fitting function, then based on the target ordinate, detect whether the target cone is in the target lane to obtain a first detection result.

[0033] The cone obstacle avoidance method provided in the embodiment of the present application determines the target ordinate and target horizontal coordinates corresponding to the target cone according to the position information of the target cone in the vehicle coordinate system, thereby ensuring the accuracy of the determined target ordinate and target horizontal coordinates. If the target ordinate exceeds the ordinate value range of the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function, and the original right-right lane line fitting function, then the output indicates that it is currently impossible to determine whether the target cone is in the target lane, thereby ensuring the accuracy of the output that it is currently impossible to determine whether the target cone is in the target lane. If the target ordinate is within the ordinate value range of the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function, and the original right-right lane line fitting function, then based on the target ordinate, it is detected whether the target cone is in the target lane to obtain a first detection result, thereby ensuring the accuracy of the obtained first detection result.

[0034] In an optional implementation, detecting whether the target cone is in the target lane according to the target longitudinal coordinate to obtain a first detection result includes:

[0035] Substitute the target ordinate into the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function and the original right-right lane line fitting function, respectively, to obtain the original left lane line abscissa, the original left-left lane line abscissa, the original right lane line abscissa and the original right-right lane abscissa;

[0036] If the target horizontal coordinate is within the horizontal coordinate range of the two lane lines corresponding to the target lane, it is determined that the target cone is in the target lane;

[0037] If the target horizontal coordinate is not within the horizontal coordinate range of the two lane lines corresponding to the target lane, it is determined that the target cone is not in the target lane.

[0038] The cone obstacle avoidance method provided in the embodiment of the present application substitutes the target ordinate into the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function and the original right-right lane line fitting function respectively, and obtains the original left lane line horizontal coordinate, the original left-left lane line horizontal coordinate, the original right lane line horizontal coordinate and the original right-right lane horizontal coordinate respectively; if the target horizontal coordinate is within the horizontal coordinate range of the two lane lines corresponding to the target lane, it is determined that the target cone is in the target lane, ensuring the accuracy of the result that the target cone is in the target lane. If the target horizontal coordinate is not within the horizontal coordinate range of the two lane lines corresponding to the target lane, it is determined that the target cone is not in the target lane, ensuring the accuracy of the result that the target cone is not in the target lane.

[0039] In an optional implementation, determining the target horizontal coordinate corresponding to the target cone barrel according to the position information of the target cone barrel includes:

[0040] Based on the position information of the target cone in the vehicle coordinate system, determine the longitudinal distance between the target cone and the target vehicle, and the current horizontal coordinate corresponding to the target cone;

[0041] According to the corresponding relationship between the longitudinal distance and the deviation threshold, the deviation threshold of the target cone barrel relative to the target vehicle is determined;

[0042] Add the current horizontal coordinate to the deviation threshold and calculate the target horizontal coordinate.

[0043] The cone barrel obstacle avoidance method provided in the embodiment of the present application determines the longitudinal distance between the target cone barrel and the target vehicle, and the current horizontal coordinate corresponding to the target cone barrel based on the position information of the target cone barrel in the vehicle coordinate system; determines the deviation threshold of the target cone barrel relative to the target vehicle according to the corresponding relationship between the longitudinal distance and the deviation threshold, thereby ensuring the accuracy of the deviation threshold of the determined target cone barrel relative to the target vehicle. The current horizontal coordinate is added with the deviation threshold to calculate the target horizontal coordinate, thereby ensuring the accuracy of the calculated target horizontal coordinate.

[0044] In an optional implementation, determining whether the target cone is in the target lane according to the first detection result and the second detection result includes:

[0045] If the first detection result indicates that the target cone is in the target lane, and the second detection result indicates that the target cone is not in the target lane, the curvature corresponding to the target lane is detected. If the curvature corresponding to the target lane is greater than a preset curvature threshold, it is determined that the target cone is not in the target lane.

[0046] If the first detection result indicates that the target cone is not in the target lane, and the second detection result indicates that the target cone is not in the target lane, it is determined that the target cone is not in the target lane;

[0047] If the first detection result indicates that the target cone is not in the target lane, and the second detection result indicates that the target cone is in the target lane, it is determined that the target cone is not in the target lane;

[0048] If the first detection result indicates that the target cone is in the target lane, and the second detection result indicates that the target cone is in the target lane, filtering the position information of the target cone in the vehicle coordinate system to generate target position information;

[0049] Based on the target position, detect whether the target cone is in the target lane.

[0050] The cone obstacle avoidance method provided in the embodiment of the present application is as follows: if the first detection result indicates that the target cone is in the target lane and the second detection result indicates that the target cone is not in the target lane, then the curvature corresponding to the target lane is detected. If the curvature corresponding to the target lane is greater than a preset curvature threshold, then it is determined that the target cone is not in the target lane, thereby ensuring the accuracy of the result that the target cone is not in the target lane. If the first detection result indicates that the target cone is not in the target lane and the second detection result indicates that the target cone is not in the target lane, then it is determined that the target cone is not in the target lane, thereby ensuring the accuracy of the result that the target cone is not in the target lane. If the first detection result indicates that the target cone is not in the target lane and the second detection result indicates that the target cone is in the target lane, then it is determined that the target cone is not in the target lane, thereby ensuring the accuracy of the result that the target cone is not in the target lane.

[0051] If the first detection result indicates that the target cone is in the target lane, and the second detection result indicates that the target cone is in the target lane, the position information of the target cone in the vehicle coordinate system is filtered to generate target position information; based on the target position, detect whether the target cone is in the target lane, thereby ensuring the accuracy of the result of detecting whether the target cone is in the target lane.

[0052] In an optional implementation, based on the target position, detecting whether the target cone is in the target lane includes:

[0053] Based on the relationship between the target position and the original lane line information, detecting whether the target cone is in the target lane, and obtaining a third detection result;

[0054] Based on the relationship between the target position and the parallel lane line information, detecting whether the target cone is in the target lane, and obtaining a fourth detection result;

[0055] If the third detection result indicates that the target cone is in the target lane, and the fourth detection result indicates that the target cone is in the target lane, the front original image is recognized, and the target cone is determined from the front original image;

[0056] Based on the position information of the target cone in the original image ahead, detect whether the target cone is in the target lane;

[0057] If so, it is determined that the target cone is in the target lane.

[0058] The cone obstacle avoidance method provided in the embodiment of the present application detects whether the target cone is in the target lane based on the relationship between the target position and the original lane line information, and obtains a third detection result; detects whether the target cone is in the target lane based on the relationship between the target position and the parallel lane line information, and obtains a fourth detection result; if the third detection result indicates that the target cone is in the target lane and the fourth detection result indicates that the target cone is in the target lane, the original image in front is recognized and the target cone is determined from the original image in front; based on the position information of the target cone in the original image in front, detects whether the target cone is in the target lane; if so, determines that the target cone is in the target lane, thereby ensuring the accuracy of the determination that the target cone is in the target lane.

[0059] In an optional implementation, the target lane includes a lane for the vehicle, a lane on the left side of the vehicle, and a lane on the right side of the vehicle. According to the detection result, the target vehicle is controlled, including:

[0060] If the detection result is that it is currently impossible to determine whether the target cone is in the lane of the vehicle, the lane on the left of the vehicle, or the lane on the right of the vehicle, or the detection result is that the target cone is not in the lane of the vehicle and it is currently impossible to determine whether the target cone is in the lane on the left of the vehicle or the lane on the right of the vehicle, then obtain the horizontal and vertical distances between the target cone and the target vehicle;

[0061] Determine whether to brake, deflect, or change lanes of the target vehicle based on the lateral and longitudinal distances;

[0062] If the detection result shows that the target cone is in the vehicle's lane but not in the left lane or the right lane of the vehicle, the target vehicle is controlled to change lanes to the left lane or the right lane of the vehicle;

[0063] If the detection result is that the target cone is not in the lane of the vehicle, but in the left lane or the right lane of the vehicle, the target vehicle is prohibited from braking.

[0064] The cone barrel obstacle avoidance method provided in the embodiment of the present application, if the detection result is that it is currently impossible to determine whether the target cone barrel is in the lane of the vehicle, the left lane of the vehicle, and the right lane of the vehicle, or the detection result is that the target cone barrel is not in the lane of the vehicle, and it is currently impossible to determine whether the target cone barrel is in the left lane of the vehicle or the right lane of the vehicle, then the horizontal and vertical distances between the target cone barrel and the target vehicle are obtained, and it is determined whether to brake, deflect, or change lanes of the target vehicle according to the horizontal and vertical distances, so as to avoid the occurrence of dangerous situations and ensure the accuracy of whether to brake, deflect, or change lanes of the target vehicle. If the detection result is that the target cone barrel is in the lane of the vehicle, but not in the left lane of the vehicle and the right lane of the vehicle, then the target vehicle is controlled to change lanes to the left lane of the vehicle or the right lane of the vehicle, ensuring the accuracy of controlling the target vehicle to change lanes to the left lane of the vehicle or the right lane of the vehicle. If the detection result is that the target cone barrel is not in the lane of the vehicle, but in the left lane of the vehicle or the right lane of the vehicle, then the target vehicle is controlled to prohibit braking, ensuring the accuracy of controlling the target vehicle to prohibit braking.

[0065] In a second aspect, the present invention provides a cone barrel obstacle avoidance device, the device comprising:

[0066] The first acquisition module is used to acquire the original front image corresponding to the target vehicle;

[0067] A determination module is used to identify the original image in front, determine the original lane line information in the original image in front, and perform parallel processing on the original lane line to generate parallel lane line information;

[0068] The second acquisition module is used to obtain the position information of the target cone barrel in the vehicle coordinate system;

[0069] A detection module, used to detect whether the target cone is in the target lane based on the original lane line information and the parallel lane line information;

[0070] The control module is used to control the target vehicle according to the detection result.

[0071] The cone barrel obstacle avoidance device provided in the embodiment of the present application obtains the original image in front of the target vehicle; identifies the original image in front, determines the original lane line information in the original image in front, and ensures the accuracy of the original lane line information in the determined original image in front. Then, the original lane line is processed in parallel to generate parallel lane line information, and the accuracy of the generated parallel lane line information is ensured. The position information of the target cone barrel in the vehicle coordinate system is obtained; based on the original lane line information and the parallel lane line information, it is detected whether the target cone barrel is in the target lane, and the accuracy of the obtained detection result is ensured, avoiding the accuracy of the obtained detection result caused by only detecting whether the target cone barrel is in the target lane based on the original lane line information. According to the detection result, the target vehicle is controlled, thereby ensuring the accuracy of the control of the target vehicle. This avoids the problem that the electronic equipment in the target vehicle misjudges that the cone barrel is in the lane or the adjacent lane, resulting in the problem of misbraking and lane change back. Therefore, the above method improves the accuracy of intelligent driving, thereby avoiding the problem that users lack confidence in intelligent driving, the usage rate is reduced, and it cannot be popularized.

[0072] In a third aspect, the present invention provides an electronic device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the cone barrel obstacle avoidance method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0073] In a fourth aspect, the present invention provides a target vehicle, which includes a vehicle body and an electronic device, and the electronic device is used to execute the cone barrel obstacle avoidance method of the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0075] Figure 1 is a schematic flow chart of a cone barrel obstacle avoidance method according to an embodiment of the present invention;

[0076] Figure 2 is a flow chart of another cone barrel obstacle avoidance method according to an embodiment of the present invention;

[0077] Figure 3 is a schematic diagram of an original lane line according to an embodiment of the present invention;

[0078] Figure 4is a flow chart of another cone barrel obstacle avoidance method according to an embodiment of the present invention;

[0079] Figure 5 is a structural block diagram of a cone barrel obstacle avoidance device according to an embodiment of the present invention;

[0080] Figure 6 is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention;

[0081] Figure 7 Schematic diagram of the structure of a target vehicle according to an embodiment of the present invention. DETAILED DESCRIPTION

[0082] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0083] With the rapid development of electrification and intelligence, automatic assisted driving technology has become one of the selling points of various car manufacturers. The purpose of assisted driving is not only to minimize the danger of car driving, but also to reduce the heavy driving tasks of users.

[0084] However, when the automatic assisted driving cruise is turned on, the vehicle system has a relatively simple judgment on the positional relationship between the front cone and the lane line, and may misjudge that the cone is in the current lane or the adjacent lane, resulting in problems such as misbraking and lane change backing, which directly leads to users' lack of confidence in smart driving, reduced usage rate, and failure to popularize it.

[0085] It should be noted that the method for avoiding cone barrel obstacles provided in the embodiment of the present application may be executed by a device for avoiding cone barrel obstacles, and the device for avoiding cone barrel obstacles may be implemented as part or all of an electronic device through software, hardware, or a combination of software and hardware, wherein the electronic device may be a server or a terminal, wherein the server in the embodiment of the present application may be a single server or a server cluster composed of multiple servers, and the terminal in the embodiment of the present application may be a smart phone, a personal computer, a tablet computer, a wearable device, an intelligent robot, or other intelligent hardware devices. In the following method embodiments, the execution subject is an electronic device as an example for explanation.

[0086] According to an embodiment of the present invention, an embodiment of a cone barrel obstacle avoidance method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0087] In this embodiment, a cone barrel obstacle avoidance method is provided, which can be used for the above-mentioned electronic equipment. Figure 1 is a flow chart of a cone barrel obstacle avoidance method according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0088] Step S101, obtaining the original front image corresponding to the target vehicle.

[0089] Specifically, the electronic device can collect images of the road area in front of the vehicle according to a preset time interval based on a front-view camera located on the front windshield to obtain multiple frames of original front images.

[0090] Step S102, identifying the front original image, determining the original lane line information in the front original image, and performing parallel processing on the original lane lines to generate parallel lane line information.

[0091] Specifically, the electronic device can pre-process the original image in front, such as graying, filtering, edge detection, etc., to enhance the lane line features in the image. Then, the electronic device can use the lane line detection model to identify the original image in front, and obtain the left and right lane line information of the lane where the vehicle is located, including the image ID and the lane line ID of each lane line. The lane line at this time is the original lane line. Then, the electronic device can parallel process the detected original lane lines to generate parallel lane lines. Then, the parallel lane lines are identified to generate parallel lane line information.

[0092] This step will be described in detail below.

[0093] Step S103, obtaining the position information of the target cone in the vehicle coordinate system.

[0094] Specifically, the electronic device can obtain the position information of the target cone barrel in the vehicle coordinate system based on the pinhole imaging principle and the front-view camera.

[0095] Step S104: Based on the original lane line information and the parallel lane line information, detect whether the target cone is in the target lane.

[0096] Specifically, the electronic device may compare the original lane line information with the position information of the target cone in the vehicle coordinate system to detect whether the target cone is in the original lane corresponding to the target lane. Then, the electronic device compares the parallel lane line information with the position information of the target cone in the vehicle coordinate system to detect whether the target cone is in the parallel lane corresponding to the target lane.

[0097] Then, based on the above detection results, determine whether the target cone is in the target lane.

[0098] This step will be described in detail below.

[0099] Step S105: controlling the target vehicle according to the detection result.

[0100] Specifically, the electronic device can determine the specific position of the target cone barrel according to the detection result, and then brake or steer the target vehicle or prohibit braking control according to the specific position of the target cone barrel.

[0101] This step will be described in detail below.

[0102] The cone barrel obstacle avoidance method provided in the embodiment of the present application obtains the original image in front of the target vehicle; identifies the original image in front, determines the original lane line information in the original image in front, and ensures the accuracy of the original lane line information in the determined original image in front. Then, the original lane line is processed in parallel to generate parallel lane line information, and the accuracy of the generated parallel lane line information is ensured. The position information of the target cone barrel in the vehicle coordinate system is obtained; based on the original lane line information and the parallel lane line information, it is detected whether the target cone barrel is in the target lane, and the accuracy of the obtained detection result is ensured, and it is avoided that only based on the original lane line information to detect whether the target cone barrel is in the target lane, resulting in the accuracy of the obtained detection result. According to the detection result, the target vehicle is controlled, thereby ensuring the accuracy of the control of the target vehicle. This avoids the problem that the electronic equipment in the target vehicle misjudges that the cone barrel is in the lane or the adjacent lane, resulting in the problem of misbraking and lane change back. Therefore, the above method improves the accuracy of intelligent driving, thereby avoiding the problem that users lack confidence in intelligent driving, the usage rate is reduced, and it cannot be popularized.

[0103] In this embodiment, a cone barrel obstacle avoidance method is provided, which can be used for the above-mentioned electronic equipment. Figure 2 is a flow chart of a cone barrel obstacle avoidance method according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0104] Step S201, obtaining the original front image corresponding to the target vehicle.

[0105] For details about this step, please refer to the above description of step S101, which will not be elaborated here.

[0106] Step S202, identifying the front original image, determining the original lane line information in the front original image, and performing parallel processing on the original lane lines to generate parallel lane line information.

[0107] Specifically, the original lane line information includes the original fitting function corresponding to each original lane line, and the parallel lane line information includes the parallel fitting function corresponding to each parallel lane line. The above step S202 may include the following steps:

[0108] Step S2021, input the front original image into the lane line detection model, identify the original lane lines in the front original image, and determine the position information and direction information corresponding to each original lane line.

[0109] Specifically, the electronic device can preprocess the original image in front, such as cropping, scaling, denoising and other operations, to improve the image quality and the accuracy of subsequent processing. Then, the electronic device uses a feature extractor such as a convolutional neural network (CNN) to extract features from the preprocessed image to obtain a feature representation of the image. Then, the electronic device inputs the feature representation into a lane detection model, which is usually based on deep learning algorithms such as Faster R-CNN, YOLO, etc. The lane detection model detects and locates the lane lines in the image, and outputs the position information of each original lane line, usually in the form of a bounding box.

[0110] It should be noted that the position information of the original lane line may be position information in vehicle coordinates with the center point of the rear axle of the target vehicle as the coordinate origin.

[0111] In addition, the electronic device can determine the direction information of the lane line by analyzing the position information of the lane line. For example, the slope or angle of the lane line can be calculated to indicate its direction. In order to distinguish different lane lines, each lane line is usually assigned a unique ID information. The ID information can be used for subsequent tracking and analysis. Finally, the lane line detection model will output the position information, direction information and ID information corresponding to each original lane line.

[0112] Step S2022: For each original lane line, output multiple lane line points corresponding to the original lane line according to the position and direction information corresponding to the original lane line.

[0113] Specifically, for each original lane line, the electronic device can select a suitable sampling interval based on the position and direction information of the original lane line. Then, the interval distance of sampling on the original lane line is determined according to the required accuracy and computing resources. Sampling along the original lane line: According to the selected sampling interval, starting from the starting point of the original lane line, sampling is performed along the original lane line according to a certain step length. Calculate the position of each sampling point: For each sampling point, the specific position coordinates of the point are calculated using a mathematical formula or algorithm based on the position and direction information of the original lane line. Then, the calculated position of each sampling point is output as a lane line point.

[0114] Step S2023: Based on the position information of each lane line point corresponding to the original lane line, an original fitting function corresponding to each lane line is fitted and generated.

[0115] Specifically, the electronic device may fit and generate an original fitting function corresponding to each lane line based on the position information of each lane line point corresponding to the original lane line.

[0116] The original fitting function may be a cubic polynomial, a quartic polynomial, or other polynomials. The embodiment of the present application does not specifically limit the original fitting function.

[0117] For example, Figure 3 The original lane line image is shown.

[0118] Among them, the original fitting function corresponding to each original lane line can be shown as follows:

[0119] Left left line Y = A1X 3 +B1X 2 +C1X+D1

[0120] Left line Y = A2X 3 +B2X 2 +C2X+D2

[0121] Right line Y = A3X 3 +B3X 2 +C3X+D3

[0122] Right line Y = A4X 3 +B4X 2 +C4X+D4

[0123] Step S2024: for any two adjacent original lane lines, determine the median lines corresponding to the two adjacent original lane lines along the vehicle travel direction according to the position information of the two adjacent original lane lines.

[0124] Specifically, for any two adjacent original lane lines, the electronic device can obtain the position information of the two adjacent original lane lines. Then, for the two adjacent original lane lines, multiple lane line points corresponding to the two adjacent original lane lines at the same horizontal position are obtained respectively, and then the distance between the two lane line points at the same horizontal position is calculated in sequence. According to the distance between the two lane line points, the middle point corresponding to the two lane line points is determined. In this way, multiple middle points are obtained, and each middle point is connected to generate the median line corresponding to the two adjacent original lane lines along the vehicle driving direction.

[0125] Optionally, in order to improve the accuracy and stability of the median, the median can be optimized. For example, a smoothing algorithm can be used to smooth the median, or a filtering algorithm can be used to remove noise and outliers.

[0126] Optionally, as the target vehicle moves, the position of the lane line may change. Therefore, the median line needs to be updated in real time to ensure its accuracy and reliability.

[0127] Step S2025, determining the position information corresponding to the multiple median points in the median line.

[0128] Specifically, the electronic device may determine a plurality of middle points connected to form a median line as a plurality of median points in the median line, and determine position information corresponding to the plurality of median points.

[0129] Step S2026: generating a median line fitting function by fitting according to the position information corresponding to each median point.

[0130] Specifically, the electronic device may generate a median line fitting function by fitting according to the position information corresponding to each median point.

[0131] The median line fitting function may be a cubic polynomial, a quartic polynomial, or other polynomials. The embodiment of the present application does not specifically limit the median line fitting function.

[0132] Step S2027: According to the lane width, the median line is offset to both sides to determine the parallel lane lines corresponding to the two adjacent original lane lines.

[0133] Specifically, the electronic device can also determine the lane width according to the position information of the original lane line, and then, according to the lane width, offset the median line to both sides by half of the lane width to determine the parallel lane lines corresponding to the two adjacent original lane lines.

[0134] Step S2028: Based on each parallel lane line, determine the parallel fitting function corresponding to each parallel lane line.

[0135] Specifically, the electronic device may determine the parallel fitting function corresponding to each parallel lane line based on each parallel lane line.

[0136] Exemplarily, the parallel fitting function corresponding to each parallel lane line can be as follows:

[0137] Left line Y = A'1X3 + B'1X2 + C'1X + D'1;

[0138] Left line Y = A'2X3+B'2X2+C'2X+D'2;

[0139] Right line Y = A′3X3+B′3X2+C′3X+D′3;

[0140] Right line Y=A′4X3+B′4X2+C′4X+D′4.

[0141] Step S203, obtaining the position information of the target cone in the vehicle coordinate system.

[0142] For details about this step, please refer to the above description of step S103, which will not be elaborated here.

[0143] Step S204: Based on the original lane line information and the parallel lane line information, detect whether the target cone is in the target lane.

[0144] For details about this step, please refer to the above description of step S104, which will not be elaborated here.

[0145] Step S205: Control the target vehicle according to the detection result.

[0146] For details about this step, please refer to the above description of step S105, which will not be elaborated here.

[0147] The cone obstacle avoidance method provided in the embodiment of the present application inputs the original image in front into the lane line detection model, identifies the original lane lines in the original image in front, determines the position information and direction information corresponding to each original lane line, and ensures the accuracy of the position information and direction information corresponding to each original lane line. For each original lane line, according to the position and direction information corresponding to the original lane line, multiple lane line points corresponding to the original lane line are output, and the accuracy of the multiple lane line points corresponding to the output original lane line is ensured. Based on the position information of each lane line point corresponding to the original lane line, the original fitting function corresponding to each lane line is fitted and generated, and the accuracy of the original fitting function corresponding to each lane line generated by fitting is ensured.

[0148] For any two adjacent original lane lines, the median line corresponding to the two adjacent original lane lines along the vehicle's driving direction is determined based on the position information of the two adjacent original lane lines, thereby ensuring the accuracy of the median line corresponding to the two adjacent original lane lines along the vehicle's driving direction. The position information corresponding to multiple median points in the median is determined; according to the position information corresponding to each median point, a median fitting function is generated by fitting, thereby ensuring the accuracy of the median fitting function generated by fitting. According to the lane width, the median line is offset to both sides to determine the parallel lane lines corresponding to the two adjacent original lane lines, thereby ensuring the accuracy of the parallel lane lines corresponding to the two adjacent original lane lines. Based on each parallel lane line, the parallel fitting function corresponding to each parallel lane line is determined, thereby ensuring the accuracy of the parallel fitting function corresponding to the determined parallel lane lines.

[0149] In this embodiment, a cone barrel obstacle avoidance method is provided, which can be used for the above-mentioned electronic equipment. Figure 3 is a flow chart of a cone barrel obstacle avoidance method according to an embodiment of the present invention. Figure 4 As shown, the process includes the following steps:

[0150] Step S301, obtaining the original front image corresponding to the target vehicle.

[0151] For details about this step, please refer to the above description of step S201, which will not be elaborated here.

[0152] Step S302, identifying the front original image, determining the original lane line information in the front original image, and performing parallel processing on the original lane lines to generate parallel lane line information.

[0153] For details about this step, please refer to the above description of step S202, which will not be elaborated here.

[0154] Step S303, obtaining the position information of the target cone in the vehicle coordinate system.

[0155] For details about this step, please refer to the above description of step S203, which will not be elaborated here.

[0156] Step S304: Based on the original lane line information and the parallel lane line information, detect whether the target cone is in the target lane.

[0157] Specifically, the target lane includes the vehicle lane, the vehicle left lane and the vehicle right lane. The above step S304 may include the following steps:

[0158] Step S3041: For each target lane, detect whether the target cone is in the target lane according to the original lane line information to obtain a first detection result.

[0159] Specifically, the original lane line includes an original left lane line, an original left-left lane line, an original right lane line, and an original right-right lane line. The original lane line information includes an original left lane line fitting function, an original left-left lane line fitting function, an original right lane line fitting function, and an original right-right lane line fitting function. The above step S3041 may include the following steps:

[0160] Step a1, determining the target vertical coordinate and the target horizontal coordinate corresponding to the target cone barrel according to the position information of the target cone barrel in the vehicle coordinate system.

[0161] Specifically, the electronic device can determine the target vertical coordinate and the target horizontal coordinate corresponding to the target cone barrel according to the position information of the target cone barrel in the vehicle coordinate system.

[0162] Specifically, the step a1 of “determining the target horizontal coordinate corresponding to the target cone barrel according to the position information of the target cone barrel” may include the following steps:

[0163] Step a11, based on the position information of the target cone in the vehicle coordinate system, determine the longitudinal distance between the target cone and the target vehicle, and the current horizontal coordinate corresponding to the target cone.

[0164] Specifically, the electronic device can determine the longitudinal distance between the target cone and the target vehicle, and the current horizontal coordinate corresponding to the target cone based on the position information of the target cone in the vehicle coordinate system.

[0165] Step a12, determining the deviation threshold of the target cone relative to the target vehicle based on the corresponding relationship between the longitudinal distance and the deviation threshold.

[0166] Specifically, the electronic device may receive the correspondence between the longitudinal distance and the deviation threshold input by the user, or may receive the correspondence between the longitudinal distance and the deviation threshold sent by other devices. The embodiment of the present application does not specifically limit the manner in which the electronic device obtains the correspondence between the longitudinal distance and the deviation threshold.

[0167] It should be noted that the correspondence between the longitudinal distance and the deviation threshold can be used to compensate for the position information of the target cone barrel in the vehicle coordinate system obtained by the electronic device. The correspondence between the longitudinal distance and the deviation threshold can be obtained based on multiple real vehicle implementations. For example, in a real situation, the target cone barrel is placed 50 cm to the left of the target lane line, and the lateral position of the target cone barrel detected by the actual electronic device is 30 cm to the left of the target lane line. Therefore, there is a problem of inaccurate lateral coordinates of the target cone barrel. Therefore, the electronic device can add the corresponding deviation threshold to the lateral coordinates of the target cone barrel to obtain the accurate lateral coordinates of the target cone barrel.

[0168] In addition, it should be noted that the correspondence between the longitudinal distance and the deviation threshold may vary depending on the model of the target vehicle, and may also vary depending on the internal parameters of the camera in the target vehicle. Therefore, the embodiment of the present application does not specifically limit the correspondence between the longitudinal distance and the deviation threshold.

[0169] Step a13, adding the current horizontal coordinate to the deviation threshold to calculate the target horizontal coordinate.

[0170] Specifically, the electronic device may add the current horizontal coordinate to the deviation threshold to calculate the target horizontal coordinate.

[0171] Step a2: If the target ordinate exceeds the ordinate value range of the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function, and the original right-right lane line fitting function, then it is output that it is currently impossible to determine whether the target cone is in the target lane.

[0172] Specifically, the electronic device can first determine the vertical coordinate value range corresponding to the original lane line based on the original lane line determined in the original image ahead, thereby determining the vertical coordinate value range of the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function and the original right-right lane line fitting function.

[0173] Then, the target ordinate is compared with the ordinate value ranges of the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function, and the original right-right lane line fitting function.

[0174] If the target ordinate exceeds the ordinate value range of the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function and the original right-right lane line fitting function, the output is that it is currently impossible to determine whether the target cone is in the target lane.

[0175] Step a3: If the target ordinate is within the ordinate value range of the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function, and the original right-right lane line fitting function, then based on the target ordinate, detect whether the target cone is in the target lane to obtain a first detection result.

[0176] Specifically, the above step a3 may include the following steps:

[0177] Step a31, substitute the target ordinate into the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function and the original right-right lane line fitting function respectively, and obtain the original left lane line horizontal coordinate, the original left-left lane line horizontal coordinate, the original right lane line horizontal coordinate and the original right-right lane horizontal coordinate respectively.

[0178] Specifically, the electronic device can substitute the target vertical coordinate into the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function and the original right-right lane line fitting function, respectively, to obtain the original left lane line horizontal coordinate, the original left-left lane line horizontal coordinate, the original right lane line horizontal coordinate and the original right-right lane horizontal coordinate.

[0179] Step a32: If the target horizontal coordinate is within the horizontal coordinate range of the two lane lines corresponding to the target lane, it is determined that the target cone is in the target lane.

[0180] Specifically, the electronic device can compare the target horizontal coordinate with the original left lane line horizontal coordinate, the original left-left lane line horizontal coordinate, the original right lane line horizontal coordinate and the original right-right lane horizontal coordinate. If the target horizontal coordinate is within the horizontal coordinate range of the two lane lines corresponding to the target lane, it is determined that the target cone is in the target lane.

[0181] Step a33: If the target horizontal coordinate is not within the horizontal coordinate range of the two lane lines corresponding to the target lane, it is determined that the target cone is not in the target lane.

[0182] Specifically, if the target horizontal coordinate is not within the horizontal coordinate range of the two lane lines corresponding to the target lane, it is determined that the target cone is not in the target lane.

[0183] Exemplarily, if the original left lane line horizontal coordinate < the target horizontal coordinate X < the original left lane line horizontal coordinate, it is determined that the target cone is in the left lane of the vehicle.

[0184] Step S3042: Detect whether the target cone is in the target lane based on the parallel lane line information to obtain a second detection result.

[0185] Specifically, the electronic device can substitute the target ordinate into the parallel left lane line fitting function, the parallel left-left lane line fitting function, the parallel right lane line fitting function and the parallel right-right lane line fitting function respectively to obtain the parallel left lane line abscissa, the parallel left-left lane line abscissa, the parallel right lane line abscissa and the parallel right-right lane abscissa.

[0186] Then, the electronic device can compare the target horizontal coordinate with the horizontal coordinate of the parallel left lane line, the horizontal coordinate of the parallel left-left lane line, the horizontal coordinate of the parallel right lane line, and the horizontal coordinate of the parallel right-right lane. If the target horizontal coordinate is within the horizontal coordinate range of the two lane lines corresponding to the target lane, it is determined that the target cone is in the target lane.

[0187] If the target horizontal coordinate is not within the horizontal coordinate range of the two lane lines corresponding to the target lane, it is determined that the target cone is not in the target lane.

[0188] Exemplarily, if the horizontal coordinate of the parallel left lane line < the target horizontal coordinate X < the horizontal coordinate of the parallel left lane line, it is determined that the target cone is in the left lane of the vehicle.

[0189] Step S3043: Determine whether the target cone is in the target lane based on the first detection result and the second detection result.

[0190] The target lane is any one of the lane, left lane and right lane corresponding to the target vehicle.

[0191] Specifically, the above step S3043 may include the following steps:

[0192] Step c1: if the first detection result indicates that the target cone is in the target lane and the second detection result indicates that the target cone is not in the target lane, the curvature corresponding to the target lane is detected. If the curvature corresponding to the target lane is greater than a preset curvature threshold, it is determined that the target cone is not in the target lane.

[0193] Specifically, if the first detection result indicates that the target cone is in the target lane, and the second detection result indicates that the target cone is not in the target lane, the electronic device can determine the curvature corresponding to the target lane through the direction information corresponding to the original lane line. If the curvature corresponding to the target lane is greater than the preset curvature threshold, the electronic device uses the result of the parallel fitting function detection, that is, the second detection result, as the basis, and therefore determines that the target cone is not in the target lane.

[0194] Step c2: If the first detection result indicates that the target cone is not in the target lane and the second detection result indicates that the target cone is not in the target lane, it is determined that the target cone is not in the target lane.

[0195] Specifically, if the first detection result indicates that the target cone is not in the target lane, and the second detection result indicates that the target cone is not in the target lane, the electronic device determines that the target cone is not in the target lane.

[0196] Step c3: if the first detection result indicates that the target cone is not in the target lane, and the second detection result indicates that the target cone is in the target lane, it is determined that the target cone is not in the target lane.

[0197] Specifically, if the first detection result indicates that the target cone is not in the target lane, and the second detection result indicates that the target cone is in the target lane, the electronic device determines that the target cone is not in the target lane.

[0198] Step c4: If the first detection result indicates that the target cone is in the target lane and the second detection result indicates that the target cone is in the target lane, the position information of the target cone in the vehicle coordinate system is filtered to generate target position information.

[0199] Specifically, if the first detection result indicates that the target cone is in the target lane and the second detection result indicates that the target cone is in the target lane, the electronic device can generate target position information by filtering the position information of the target cone in the vehicle coordinate system using a Mann filter based on the historical position information of the target cone in the vehicle coordinate system.

[0200] Step c5, based on the target position, detect whether the target cone is in the target lane.

[0201] Specifically, the above step c5 may include the following steps:

[0202] Step c51, based on the relationship between the target position and the original lane line information, detect whether the target cone is in the target lane to obtain a third detection result.

[0203] Specifically, the electronic device can detect whether the target cone is in the target lane based on the relationship between the target position and the original lane line information, and obtain a third detection result.

[0204] The specific implementation method for obtaining the third detection result may refer to the specific implementation method for obtaining the first detection result, which will not be described in detail here.

[0205] Step c52, based on the relationship between the target position and the parallel lane line information, detect whether the target cone is in the target lane to obtain a fourth detection result.

[0206] Specifically, the specific implementation method for obtaining the fourth detection result can refer to the specific implementation method for obtaining the second detection result mentioned above, which will not be described in detail here.

[0207] Step c53: If the third detection result indicates that the target cone is in the target lane and the fourth detection result indicates that the target cone is in the target lane, the front original image is identified and the target cone is determined from the front original image.

[0208] Specifically, if the third detection result indicates that the target cone is in the target lane and the fourth detection result indicates that the target cone is in the target lane, the electronic device can use the target detection model to identify the original image in front and determine the target cone from the original image in front.

[0209] Step c54, based on the position information of the target cone in the original image ahead, detect whether the target cone is in the target lane.

[0210] Specifically, the electronic device detects whether the target cone is in the target lane based on the pixel position of the target cone in the front original image.

[0211] Step c55: If yes, determine that the target cone is in the target lane.

[0212] Specifically, if the target cone is in the target lane, the electronic device determines that the target cone is in the target lane.

[0213] Optionally, if the third detection result indicates that the target cone is in the target lane, and the fourth detection result indicates that the target cone is not in the target lane, a curvature corresponding to the target lane is detected, and if the curvature corresponding to the target lane is greater than a preset curvature threshold, it is determined that the target cone is not in the target lane;

[0214] If the third detection result indicates that the target cone is not in the target lane, and the fourth detection result indicates that the target cone is not in the target lane, it is determined that the target cone is not in the target lane;

[0215] If the third detection result indicates that the target cone is not in the target lane, and the fourth detection result indicates that the target cone is in the target lane, it is determined that the target cone is not in the target lane.

[0216] Exemplarily, the electronic device substitutes the target ordinate Y corresponding to the target cone into the original fitting function corresponding to each lane line, i.e., Y=f(x), and the parallel fitting function corresponding to each parallel lane line, i.e., Y=f′(x) function, to obtain left-left line X, left line X, right line X, right-right line X, left-left line X′, left line X′, right line X′, right-right line X′.

[0217] Then, the electronic device compares the target horizontal coordinate X corresponding to the target cone with the X value obtained by the above curve equation, and assigns Unknown, Yes, or No flags. The specific strategy is as follows:

[0218] (1) When the target ordinate Y corresponding to the target cone exceeds the value range of the original fitting function Y=f(x) corresponding to each lane line and the parallel fitting function Y=f′(x) corresponding to each parallel lane line, the electronic device directly outputs the Unknown flag; that is, it outputs that it is currently impossible to determine whether the target cone is in the target lane.

[0219] (2) When the target ordinate Y corresponding to the target cone is within the value range of the original fitting function Y=f(x) corresponding to each lane line and the parallel fitting function Y=f′(x) corresponding to each parallel lane line, substitute the target ordinate X corresponding to the target cone into the judgment of the lateral position, mainly based on the original fitting function Y=f(x) corresponding to each lane line. Then, make three judgments on whether the target cone is in the target lane:

[0220] 1) If the first detection result indicates that the target cone is in the target lane, and the second detection result indicates that the target cone is not in the target lane, if the curvature corresponding to the target lane is greater than the preset curvature threshold, the second detection result shall prevail, and it shall be determined that the target cone is not in the target lane, and the target lane shall be assigned a NO flag position;

[0221] Exemplarily, if the original left lane line horizontal coordinate < target horizontal coordinate X < original left lane line horizontal coordinate, it is determined that the target cone is in the left lane of the vehicle, that is, the first detection result indicates that the target cone is in the left lane of the vehicle.

[0222] 2) If the first detection result indicates that the target cone is not in the target lane, and the second detection result indicates that the target cone is not in the target lane, it is determined that the target cone is not in the target lane, and the target lane is directly assigned a NO flag.

[0223] 3) If the first detection result indicates that the target cone is not in the target lane, and the second detection result indicates that the target cone is in the target lane, it is determined that the target cone is not in the target lane, and the target lane is directly assigned a NO flag position;

[0224] 4) If the first detection result indicates that the target cone is in the target lane, and the second detection result indicates that the target cone is in the target lane, the electronic device needs to use the position information of the target cone in the vehicle coordinate system processed by the Kalman filter to generate the target position information (target horizontal coordinate X filter, target vertical coordinate Y filter) and re-substitute it into the original fitting function corresponding to each lane line and the parallel fitting function corresponding to each parallel lane line, and then judge the lateral position to obtain the third detection result and the fourth detection result. If the above 1), 2), and 3) situations occur, the target lane is assigned a NO flag. If the third detection result indicates that the target cone is in the target lane, and the fourth detection result indicates that the target cone is in the target lane, the electronic device then identifies the original image in front and determines the target cone from the original image in front; based on the position information of the target cone in the original image in front, detect whether the target cone is in the target lane; if so, determine that the target cone is in the target lane and assign the target lane a YES flag, otherwise, assign the target lane a NO flag.

[0225] For example: For example, in the raw line equation Y=f(x), the left-left line X<target horizontal coordinate X<left line X, and in the road model line Y=f′(x), the left-left line X′<target horizontal coordinate X<left line X′, indicating that the cone barrel has not invaded the lane and the right lane. The flags of the lane and the right lane can be directly assigned to NO, but whether it invades the left lane still needs to be judged by the filtered cone barrel coordinates. If the left-left line X<Xfilter<left line X in the raw line equation Y=f(x) and the left-left line X′<Xfilter<left line X′ in the road model line Y=f′(x), the Contour point of the lane line is used to make another judgment. If it is still judged as an invasion, the corresponding lane is assigned a YES flag, otherwise the corresponding lane is assigned a NO flag.

[0226] After the flag bit given by the J3 chip inside the front camera assembly is transmitted to the MCU chip, it is output on the CAN bus as 0x0Unknown (unknown), 0x1 intruding into the current lane, 0x2 intruding into the left lane, 0x3 intruding into the right lane, and 0x4 not in the current lane.

[0227] Step S305: Control the target vehicle according to the detection result.

[0228] In an optional implementation of the present application, the electronic device may obtain detection results corresponding to multiple frames of front original images, and if the detection results corresponding to the front original images with a preset proportion in the multiple frames of front original images are consistent, the detection results corresponding to the front original images with the preset proportion are determined as the target detection results. The target vehicle is controlled according to the target detection results.

[0229] In an optional implementation manner of the present application, the above step S305 may include the following steps:

[0230] Step S3051, if the detection result is that it is currently impossible to determine whether the target cone is in the lane of the vehicle, the left lane of the vehicle, or the right lane of the vehicle, or the detection result is that the target cone is not in the lane of the vehicle and it is currently impossible to determine whether the target cone is in the left lane of the vehicle or the right lane of the vehicle, then obtain the lateral and longitudinal distances between the target cone and the target vehicle.

[0231] Specifically, if the detection result is that it is currently impossible to determine whether the target cone is in the lane of the vehicle, the left lane of the vehicle, and the right lane of the vehicle, or the detection result is that the target cone is not in the lane of the vehicle and it is currently impossible to determine whether the target cone is in the left lane of the vehicle or the right lane of the vehicle, the electronic device obtains the lateral and longitudinal distances between the target cone and the target vehicle.

[0232] Step S3052, judging whether to brake, deflect or change lanes of the target vehicle according to the lateral and longitudinal distances.

[0233] Specifically, the electronic device can compare the transverse and longitudinal distances between the target cone and the target vehicle with a preset safety distance threshold. If the transverse and longitudinal distances are less than the safety distance threshold, it is considered that there is a risk of collision. Specifically, if the transverse and longitudinal distances are too close and the target vehicle is moving at a high speed, braking measures may be required to reduce the speed to avoid collision. If the transverse and longitudinal distances are close but the target vehicle is moving at a low speed, it may be considered to avoid the cone by offsetting to maintain a safe distance.

[0234] In some cases, if there are cones in the current lane and other lanes are safe, it may be necessary to change lanes to avoid a collision with the cones.

[0235] Step S3053: If the detection result shows that the target cone is in the lane of the vehicle but not in the left lane or the right lane of the vehicle, the target vehicle is controlled to change lanes to the left lane or the right lane of the vehicle.

[0236] Specifically, if the detection result is that the target cone is in the lane of the vehicle but not in the left lane or the right lane of the vehicle, the electronic device controls the target vehicle to change lanes to the left lane or the right lane of the vehicle.

[0237] Step S3054: If the detection result shows that the target cone is not in the lane of the vehicle, but in the left lane or the right lane of the vehicle, the target vehicle is controlled to prohibit braking.

[0238] Specifically, if the detection result is that the target cone is not in the lane of the vehicle, but in the left lane of the vehicle or the right lane of the vehicle, the electronic device controls the target vehicle to prohibit braking.

[0239] Exemplarily, the detailed signal of the cone barrel flag is as follows:

[0240]

[0241]

[0242] Specifically, for driving in this lane: for "0x0 Unknown", the control end determines whether to brake or change lanes based on the lateral and longitudinal distances of the cone barrel; for "0x1 intrusion into this lane", braking or lane changing is performed; for "0x2 intrusion into the left lane", "0x3 intrusion into the right lane", and "0x4 not in this lane", no braking is performed.

[0243] Lane change:

[0244] Depending on whether the corresponding lane reference cone flag is "0x2 intruding left lane" or "0x3 intruding right lane", if so, lane changing is prohibited; if it is "0x0 UNKOWN" or "0x4 not in this lane", the control end determines whether to change lanes based on the horizontal and vertical distances of the cone.

[0245] The cone obstacle avoidance method provided in the embodiment of the present application determines, for each target lane, the target longitudinal coordinate corresponding to the target cone according to the position information of the target cone in the vehicle coordinate system. Based on the position information of the target cone in the vehicle coordinate system, the longitudinal distance between the target cone and the target vehicle and the current horizontal coordinate corresponding to the target cone are determined; based on the correspondence between the longitudinal distance and the deviation threshold, the deviation threshold of the target cone relative to the target vehicle is determined, thereby ensuring the accuracy of the deviation threshold of the determined target cone relative to the target vehicle. The current horizontal coordinate is added with the deviation threshold to calculate the target horizontal coordinate, thereby ensuring the accuracy of the calculated target horizontal coordinate.

[0246] If the target ordinate exceeds the ordinate value range of the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function and the original right-right lane line fitting function, then the output is that it is currently impossible to determine whether the target cone barrel is in the target lane, ensuring the accuracy of the output that it is currently impossible to determine whether the target cone barrel is in the target lane.

[0247] If the target ordinate is within the ordinate value range of the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function, and the original right-right lane line fitting function, the target ordinate is substituted into the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function, and the original right-right lane line fitting function, respectively, to obtain the original left lane line horizontal coordinate, the original left-left lane line horizontal coordinate, the original right lane line horizontal coordinate, and the original right-right lane horizontal coordinate; if the target horizontal coordinate is within the horizontal coordinate range of the two lane lines corresponding to the target lane, it is determined that the target cone is in the target lane, ensuring the accuracy of the result that the target cone is in the target lane. If the target horizontal coordinate is not within the horizontal coordinate range of the two lane lines corresponding to the target lane, it is determined that the target cone is not in the target lane, ensuring the accuracy of the result that the target cone is not in the target lane.

[0248] According to the parallel lane line information, it is detected whether the target cone barrel is in the target lane, and a second detection result is obtained, thereby ensuring the accuracy of the obtained second detection result.

[0249] If the first detection result indicates that the target cone is in the target lane, and the second detection result indicates that the target cone is not in the target lane, the curvature corresponding to the target lane is detected. If the curvature corresponding to the target lane is greater than the preset curvature threshold, it is determined that the target cone is not in the target lane, thereby ensuring the accuracy of the result that the target cone is not in the target lane. If the first detection result indicates that the target cone is not in the target lane, and the second detection result indicates that the target cone is not in the target lane, it is determined that the target cone is not in the target lane, thereby ensuring the accuracy of the result that the target cone is not in the target lane. If the first detection result indicates that the target cone is not in the target lane, and the second detection result indicates that the target cone is in the target lane, it is determined that the target cone is not in the target lane, thereby ensuring the accuracy of the result that the target cone is not in the target lane.

[0250] If the first detection result indicates that the target cone is in the target lane, and the second detection result indicates that the target cone is in the target lane, the position information of the target cone in the vehicle coordinate system is filtered to generate the target position information. Based on the relationship between the target position and the original lane line information, it is detected whether the target cone is in the target lane to obtain the third detection result; based on the relationship between the target position and the parallel lane line information, it is detected whether the target cone is in the target lane to obtain the fourth detection result; if the third detection result indicates that the target cone is in the target lane, and the fourth detection result indicates that the target cone is in the target lane, the original image in front is identified, and the target cone is determined from the original image in front; based on the position information of the target cone in the original image in front, it is detected whether the target cone is in the target lane; if so, it is determined that the target cone is in the target lane, ensuring the accuracy of the target cone being in the target lane.

[0251] If the detection result is that it is currently impossible to determine whether the target cone is in the lane of the vehicle, the left lane of the vehicle, or the right lane of the vehicle, or the detection result is that the target cone is not in the lane of the vehicle, and it is currently impossible to determine whether the target cone is in the left lane of the vehicle or the right lane of the vehicle, then the horizontal and vertical distances between the target cone and the target vehicle are obtained, and it is determined whether to brake, deflect, or change lanes of the target vehicle based on the horizontal and vertical distances, so as to avoid dangerous situations and ensure the accuracy of whether to brake, deflect, or change lanes of the target vehicle. If the detection result is that the target cone is in the lane of the vehicle, but not in the left lane of the vehicle or the right lane of the vehicle, then the target vehicle is controlled to change lanes to the left lane of the vehicle or the right lane of the vehicle, ensuring the accuracy of controlling the target vehicle to change lanes to the left lane of the vehicle or the right lane of the vehicle. If the detection result is that the target cone is not in the lane of the vehicle, but in the left lane of the vehicle or the right lane of the vehicle, then the target vehicle is prohibited from braking, ensuring the accuracy of controlling the target vehicle to prohibit braking.

[0252] In this embodiment, a cone barrel obstacle avoidance device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0253] This embodiment provides a cone barrel obstacle avoidance device, such as Figure 5 As shown, including:

[0254] The first acquisition module 401 is used to acquire the original front image corresponding to the target vehicle;

[0255] The determination module 402 is used to identify the original image in front, determine the original lane line information in the original image in front, and perform parallel processing on the original lane line to generate parallel lane line information;

[0256] The second acquisition module 403 is used to obtain the position information of the target cone barrel in the vehicle coordinate system;

[0257] A detection module 404 is used to detect whether the target cone is in the target lane based on the original lane line information and the parallel lane line information;

[0258] The control module 405 is used to control the target vehicle according to the detection result.

[0259] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0260] The cone barrel obstacle avoidance device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0261] The embodiment of the present invention also provides an electronic device having the above Figure 6 The cone barrel obstacle avoidance device shown.

[0262] See also Figure 6 , Figure 6 is a schematic diagram of the structure of an electronic device provided by an optional embodiment of the present invention, such as Figure 6 As shown, the electronic device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the electronic device, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple electronic devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 6 A processor 10 is taken as an example.

[0263] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0264] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0265] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0266] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0267] The electronic device further comprises a communication interface 30 for the electronic device to communicate with other devices or a communication network.

[0268] The present application embodiment provides a target vehicle, such as Figure 7 As shown, the target vehicle includes a vehicle body and an electronic device, and the electronic device is used to execute the cone barrel obstacle avoidance method of any of the above-mentioned embodiments.

[0269] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0270] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0271] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A cone barrel obstacle avoidance method, characterized in that: The method comprises: Obtain the original front image corresponding to the target vehicle; Identify the front original image, determine original lane line information in the front original image, and perform parallel processing on the original lane lines to generate parallel lane line information; Get the position information of the target cone in the vehicle coordinate system; Based on the original lane line information and the parallel lane line information, detecting whether the target cone is in a target lane; The target vehicle is controlled according to the detection result.

2. The method according to claim 1, characterized in that The original lane line information includes an original fitting function corresponding to each original lane line, and the identifying the front original image to determine the original lane line information in the front original image includes: Input the front original image into the lane line detection model, identify the original lane lines in the front original image, and determine the position information and direction information corresponding to each original lane line. For each of the original lane lines, output a plurality of lane line points corresponding to the original lane line according to the position and direction information corresponding to the original lane line; Based on the position information of each lane line point corresponding to the original lane line, an original fitting function corresponding to each lane line is generated by fitting.

3. The method according to claim 2, characterized in that The parallel lane line information includes parallel fitting functions corresponding to each parallel lane line, and the parallel processing of the original lane lines to generate the parallel lane line information includes: For any two adjacent original lane lines, determine the median lines corresponding to the two adjacent original lane lines along the vehicle driving direction according to the position information of the two adjacent original lane lines; Determine position information corresponding to a plurality of median points in the median line; According to the position information corresponding to each of the median points, a median line fitting function is generated by fitting; According to the lane width, the median line is shifted to both sides to determine parallel lane lines corresponding to two adjacent original lane lines; Based on each of the parallel lane lines, a parallel fitting function corresponding to each of the parallel lane lines is determined.

4. The method according to claim 1, characterized in that The target lane includes a host vehicle lane, a left lane of the host vehicle, and a right lane of the host vehicle. The detecting whether the target cone is in the target lane based on the original lane line information and the parallel lane line information includes: For each of the target lanes, detecting whether the target cone is in the target lane according to the original lane line information, and obtaining a first detection result; According to the parallel lane line information, detecting whether the target cone is in the target lane to obtain a second detection result; According to the first detection result and the second detection result, it is determined whether the target cone is in the target lane, where the target lane is any one of the lane, the left lane, and the right lane corresponding to the target vehicle.

5. The method according to claim 4, characterized in that The original lane line includes an original left lane line, an original left-left lane line, an original right lane line, and an original right-right lane line. The original lane line information includes an original left lane line fitting function, an original left-left lane line fitting function, an original right lane line fitting function, and an original right-right lane line fitting function. According to the original lane line information, detecting whether the target cone is in the target lane to obtain a first detection result includes: Determine the target ordinate and target abscissa corresponding to the target cone according to the position information of the target cone in the vehicle coordinate system; If the target ordinate exceeds the ordinate value range of the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function and the original right-right lane line fitting function, then outputting that it is currently impossible to determine whether the target cone is in the target lane; If the target ordinate is within the ordinate value range of the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function and the original right-right lane line fitting function, then based on the target ordinate, detect whether the target cone is in the target lane to obtain a first detection result.

6. The method according to claim 5, characterized in that The detecting, according to the target longitudinal coordinate, whether the target cone is in the target lane to obtain a first detection result includes: Substituting the target ordinate into the original left lane line fitting function, the original left-left lane line fitting function, the original right lane line fitting function and the original right-right lane line fitting function, respectively, to obtain the original left lane line abscissa, the original left-left lane line abscissa, the original right lane line abscissa and the original right-right lane abscissa; If the target horizontal coordinate is within the horizontal coordinate range of two lane lines corresponding to the target lane, it is determined that the target cone is in the target lane; If the target horizontal coordinate is not within the horizontal coordinate range of the two lane lines corresponding to the target lane, it is determined that the target cone barrel is not in the target lane.

7. The method according to claim 4, characterized in that Determining the target horizontal coordinate corresponding to the target cone barrel according to the position information of the target cone barrel includes: Based on the position information of the target cone in the vehicle coordinate system, determine the longitudinal distance between the target cone and the target vehicle, and the current horizontal coordinate corresponding to the target cone; Determining a deviation threshold of the target cone relative to the target vehicle according to a corresponding relationship between the longitudinal distance and the deviation threshold; The current horizontal coordinate is added to the deviation threshold to calculate the target horizontal coordinate.

8. The method according to claim 4, characterized in that The determining, according to the first detection result and the second detection result, whether the target cone is in the target lane includes: If the first detection result indicates that the target cone is in the target lane, and the second detection result indicates that the target cone is not in the target lane, then detecting the curvature corresponding to the target lane, and if the curvature corresponding to the target lane is greater than a preset curvature threshold, determining that the target cone is not in the target lane; If the first detection result indicates that the target cone is not in the target lane, and the second detection result indicates that the target cone is not in the target lane, it is determined that the target cone is not in the target lane; If the first detection result indicates that the target cone is not in the target lane, and the second detection result indicates that the target cone is in the target lane, it is determined that the target cone is not in the target lane; If the first detection result indicates that the target cone is in the target lane, and the second detection result indicates that the target cone is in the target lane, filtering the position information of the target cone in the vehicle coordinate system to generate target position information; Based on the target position, detect whether the target cone is in the target lane.

9. The method according to claim 8, characterized in that The detecting, based on the target position, whether the target cone is in the target lane includes: Based on the relationship between the target position and the original lane line information, detecting whether the target cone is in the target lane to obtain a third detection result; Based on the relationship between the target position and the parallel lane line information, detecting whether the target cone is in the target lane to obtain a fourth detection result; If the third detection result indicates that the target cone is in the target lane, and the fourth detection result indicates that the target cone is in the target lane, identifying the front original image, and determining the target cone from the front original image; Based on the position information of the target cone barrel in the front original image, detecting whether the target cone barrel is in the target lane; If so, it is determined that the target cone is in the target lane.

10. The method according to claim 1, characterized in that The target lane includes the vehicle lane, the left lane of the vehicle and the right lane of the vehicle. The target vehicle is controlled according to the detection result, including: If the detection result is that it is currently impossible to determine whether the target cone is in the lane of the vehicle, the left lane of the vehicle, and the right lane of the vehicle, or the detection result is that the target cone is not in the lane of the vehicle and it is currently impossible to determine whether the target cone is in the left lane of the vehicle or the right lane of the vehicle, then the lateral and longitudinal distances between the target cone and the target vehicle are obtained; Determining whether to brake, deflect, or change lanes of the target vehicle according to the lateral and longitudinal distances; If the detection result shows that the target cone is in the lane of the vehicle but not in the left lane or the right lane of the vehicle, the target vehicle is controlled to change lanes to the left lane or the right lane of the vehicle; If the detection result is that the target cone is not in the lane of the vehicle, but in the left lane of the vehicle or the right lane of the vehicle, the target vehicle is controlled to prohibit braking.

11. A cone barrel obstacle avoidance device, characterized in that: The device comprises: The first acquisition module is used to acquire the original front image corresponding to the target vehicle; A determination module, used to identify the front original image, determine the original lane line information in the front original image, and perform parallel processing on the original lane line to generate parallel lane line information; The second acquisition module is used to obtain the position information of the target cone barrel in the vehicle coordinate system; A detection module, configured to detect whether the target cone is in a target lane based on the original lane line information and the parallel lane line information; The control module is used to control the target vehicle according to the detection result.

12. An electronic device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the cone barrel obstacle avoidance method according to any one of claims 1 to 10 by executing the computer instructions.

13. A target vehicle, characterized in that: The target vehicle includes a vehicle body and an electronic device, and the electronic device is used to execute the cone barrel obstacle avoidance method described in any one of claims 1 to 10.

Images