Vehicle active obstacle avoidance control method and system, storage medium and terminal
Through the improved YOLOv8 network model and vehicle-mounted lidar technology, obstacle-avoiding vehicles are screened and identified, obstacle-avoiding distance is calculated, and the collision risk caused by distraction is solved, and the accuracy and safety of vehicle obstacle-avoiding is achieved.
Patent Information
- Application Number
- CN202411266773.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-12-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In complex traffic environments, drivers may cause obstacle avoidance judgments due to fatigue or distraction, increasing the risk of collision, especially on multi-lane expressways.
The improved YOLOv8 network model is used to detect and identify traffic images, and the barrier vehicles on the front side, the front side and the back side lane are selected. The vehicle distance and speed are obtained through the vehicle-mounted lidar, and the extreme obstacle avoidance distance and early warning avoidance are calculated. Obstacle distance, risk warning and active deceleration control.
It improves the accuracy and safety of vehicle obstacle avoidance, reduces the detection range and calculation amount of lidar, and reduces the risk of collision.
Smart Images

Figure CN119078814A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of automobile assisted driving technology, and in particular to a vehicle active obstacle avoidance control method and system, a storage medium and a terminal. Background Art
[0002] Assisted driving refers to a technology that uses the collaborative efforts of artificial intelligence, visual computing, radar, monitoring devices, and global positioning systems to allow the controller to assist in operating the vehicle to ensure the safe driving of the car on the road.
[0003] At present, cars are driving fast on the road, and drivers need to face complex and changing traffic environments. Drivers are prone to fatigue and lack of concentration during driving. If an emergency occurs, it is too late for the driver to take emergency action by the time he finds out, and there may be a risk of collision. Especially on multi-lane expressways with a large number of vehicles, drivers are prone to make misjudgments when changing lanes to avoid obstacles, which may lead to traffic accidents. Summary of the Invention
[0004] In order to solve the above technical problems, the present invention provides a vehicle active obstacle avoidance control method and system, a storage medium and a terminal.
[0005] The present invention adopts the following technical solution: a vehicle active obstacle avoidance control method, the method comprising:
[0006] Acquire a traffic image of a target vehicle, detect and identify the traffic image based on a pre-trained improved YOLOv8 network model, and obtain multiple vehicles to be tested in the traffic image;
[0007] Obtaining the position information of the vehicle to be tested, and screening out the vehicle with a front obstacle in the same lane, the vehicle with a front obstacle in the side lane, and the vehicle with a rear obstacle in the side lane from the vehicles to be tested according to the position information;
[0008] Based on the vehicle-mounted laser radar, the current distances of the target vehicle to the obstacle vehicle in front of the same lane, the obstacle vehicle in front of the side lane, and the obstacle vehicle in the rear of the side lane are obtained, and the current speeds of the obstacle vehicle in front of the same lane and the obstacle vehicle in front of the side lane are obtained;
[0009] Obtaining a limit obstacle avoidance distance and a warning obstacle avoidance distance based on a current distance from the target vehicle to the obstacle vehicle in front of the same lane and a current speed of the obstacle vehicle in front of the same lane;
[0010] If the current distance between the target vehicle and the obstacle vehicle ahead in the same lane is greater than or equal to the warning obstacle avoidance distance, no warning prompt is issued; if the current distance between the target vehicle and the obstacle vehicle ahead in the same lane is less than or equal to the limit obstacle avoidance distance, a collision prompt is issued and the target vehicle is controlled to brake and decelerate;
[0011] If the current distance between the target vehicle and the vehicle with the front obstacle in the same lane is between the warning obstacle avoidance distance and the limit obstacle avoidance distance, an obstacle avoidance warning is issued. At the same time, based on the current speeds of the vehicle with the front obstacle in the same lane and the vehicle with the front obstacle in the side lane, and the current distances between the target vehicle and the vehicle with the front obstacle in the same lane, the vehicle with the front obstacle in the side lane, and the vehicle with the rear obstacle in the side lane, it is determined whether to issue a lane change prompt.
[0012] The vehicle active obstacle avoidance control method of the present invention detects and identifies traffic images by adopting an improved YOLOv8 network model, and can obtain the detailed distribution of vehicles to be tested on the lane, avoiding the situation where radar causes incorrect identification of obstacle vehicles, and then screening the vehicles to be tested, removing vehicles to be tested that do not affect the driving safety of the target vehicle, and only retaining the obstacle vehicles in the front of the same lane, the obstacle vehicles in the front of the side lane, and the obstacle vehicles in the rear of the side lane. The three vehicles to be tested are used as obstacles for radar ranging and speed measurement, which effectively reduces the detection range of the laser radar, reduces the amount of calculation, and improves the response efficiency of the vehicle-computer system, making vehicle obstacle avoidance more accurate and safe. By calculating the limit obstacle avoidance distance and the warning obstacle avoidance distance, and then referring to the driving status of the obstacle vehicles in the front of the same lane, the obstacle vehicles in the front of the side lane, and the obstacle vehicles in the rear of the side lane, the driver is given a risk warning or the target vehicle is actively controlled to slow down, avoiding the situation where the driver causes a collision due to distracted driving, ensuring the safety of the target vehicle in obstacle avoidance, and reducing the collision risk.
[0013] Furthermore, the improved YOLOv8 network model includes a backbone network, a neck network and a detection layer. A GAM attention mechanism module is added before the neck network is connected to the medium-sized detection head and the large-sized detection head of the detection layer. A GSBottleneck lightweight bottleneck layer is introduced based on the GSConv lightweight convolution. A one-time aggregation method is used to construct a VoV-GSCSP cross-level partial network module, and the VoV-GSCSP cross-level partial network module is used to replace the original C2f module of the neck network. The regression loss function of the improved YOLOv8 network model is the WIoU_v3 loss function.
[0014] Furthermore, the WIoU_v3 loss function is:
[0015] P WIoU_v3 =θ×k WIoU ×p IoU
[0016]
[0017]
[0018] Among them, P WI oU_v3 is the WIoU_v3 loss function, θ is the non-monotonic focusing coefficient, k WIoU is the normalized distance between the center point of the real box and the predicted box, p IoU is the loss value of the bounding box IoU, x is the horizontal coordinate of the center point of the real box, y is the vertical coordinate of the center point of the real box, and x j is the horizontal coordinate of the center point of the prediction box, y j is the vertical coordinate of the center point of the prediction box, w j is the width of the minimum prediction box, h j is the height of the minimum prediction box, ω is the outlier degree, is the gradient gain coefficient, and i is a constant.
[0019] Furthermore, the step of obtaining the position information of the vehicle to be tested and screening out the vehicle with a front obstacle in the same lane, the vehicle with a front obstacle in the side lane, and the vehicle with a rear obstacle in the side lane from the vehicles to be tested according to the position information specifically includes:
[0020] A two-dimensional coordinate system is established on the traffic image with the target vehicle as the origin, and a coordinate point is defined for the vehicle to be tested on the traffic image, with the coordinate point of the vehicle to be tested serving as the position information; wherein the horizontal axis of the two-dimensional coordinate system is perpendicular to the lane, the vertical axis is parallel to the lane, the coordinate point of the vehicle to be tested is the center point of the identification frame of the vehicle to be tested, and the origin is on the centerline of the current lane;
[0021] According to the width of the lane and the coordinate point of the vehicle to be tested, the vehicle to be tested that belongs to the same lane as the target vehicle is defined as a same-lane vehicle, and the vehicle to be tested that does not belong to the same lane as the target vehicle is defined as a side-lane vehicle;
[0022] The vehicle to be tested with the smallest positive vertical coordinate among the vehicles in the same lane is defined as the front obstacle vehicle in the same lane, the vehicle to be tested with the smallest positive vertical coordinate among the vehicles in the side lane is defined as the front obstacle vehicle in the side lane, and the vehicle to be tested with the largest vertical coordinate less than or equal to zero among the vehicles in the side lane is defined as the rear obstacle vehicle in the side lane.
[0023] Furthermore, the limit obstacle avoidance distance is:
[0024]
[0025] Among them, L e is the limit obstacle avoidance distance, v a is the current speed of the target vehicle a, v b is the current speed of the obstacle vehicle b in front of the same lane, t1 is the driver's reaction time, t2 is the response delay of the vehicle computer, μ is the road adhesion coefficient, g is the acceleration of gravity, l abThe current distance between the target vehicle a and the obstacle vehicle b in the same lane is when the current distance between the target vehicle and the obstacle vehicle in front of the target vehicle is less than or equal to the limit obstacle avoidance distance and the target vehicle brakes and decelerates until its speed matches that of the obstacle vehicle in front of the target vehicle.
[0026] Furthermore, the warning obstacle avoidance distance is:
[0027]
[0028] Among them, t3 is the driver's warning reaction time, t3 = 2s.
[0029] Furthermore, if the current distance between the target vehicle and the vehicle with the front obstacle in the same lane is between the warning obstacle avoidance distance and the limit obstacle avoidance distance, an obstacle avoidance warning is issued, and at the same time, according to the current speeds of the vehicle with the front obstacle in the same lane and the vehicle with the front obstacle in the side lane, and the current distances from the target vehicle to the vehicle with the front obstacle in the same lane, the vehicle with the front obstacle in the side lane, and the vehicle with the rear obstacle in the side lane, the steps of determining whether to issue a lane change prompt specifically include:
[0030] If the current distance between the target vehicle and the obstacle vehicle ahead in the same lane is between the warning obstacle avoidance distance and the limit obstacle avoidance distance, an obstacle avoidance warning is issued; and if either the first constraint condition or the second constraint condition is met at the same time, the vehicle computer issues a lane change prompt;
[0031] The first constraint condition is:
[0032] v a >v c >v b
[0033] (L ac >L f )U(L ad >L e )
[0034] The second constraint is:
[0035] v b <v a <v c
[0036] (L ac >L e )U(L ad >L e )
[0037] Among them, v c is the current speed of the vehicle c with the obstacle in front of the side road, L acis the current distance from the target vehicle a to the obstacle vehicle c in front of the side road, L ad is the current distance from the target vehicle a to the obstacle vehicle d behind the side road.
[0038] The present invention also provides a vehicle active obstacle avoidance control system, the system comprising:
[0039] A first acquisition module is used to acquire a traffic image in front of a target vehicle, detect and identify the traffic image based on a pre-trained improved YOL0v8 network model, and obtain multiple vehicles to be tested in the traffic image;
[0040] a screening module, configured to obtain the position information of the vehicle to be tested, and screen out the vehicles with obstacles in front of the same lane, the vehicles with obstacles in front of the side lane, and the vehicles with obstacles in the rear of the side lane from the vehicles to be tested according to the position information;
[0041] The second acquisition module is used to obtain the current distances of the target vehicle to the obstacle vehicle in front of the same lane, the obstacle vehicle in front of the side lane, and the obstacle vehicle in the rear of the side lane based on the vehicle-mounted laser radar, and simultaneously obtain the current speeds of the obstacle vehicle in front of the same lane and the obstacle vehicle in front of the side lane;
[0042] a calculation module, configured to obtain a limit obstacle avoidance distance and a warning obstacle avoidance distance based on a current distance from the target vehicle to the obstacle vehicle in front of the same lane and a current speed of the obstacle vehicle in front of the same lane;
[0043] A first judgment module is configured to: if the current distance between the target vehicle and the obstacle vehicle in front of the same lane is greater than or equal to the warning obstacle avoidance distance, the target vehicle maintains the driving state; if the current distance between the target vehicle and the obstacle vehicle in front of the same lane is less than or equal to the limit obstacle avoidance distance, the target vehicle brakes and decelerates;
[0044] The second judgment module is used to judge whether the target vehicle changes lanes based on the current speeds of the same-lane front obstacle vehicle and the side-lane front obstacle vehicle, and the current distances from the target vehicle to the same-lane front obstacle vehicle, the side-lane front obstacle vehicle, and the side-lane rear obstacle vehicle, respectively, if the current distance from the target vehicle to the same-lane front obstacle vehicle is between the warning obstacle avoidance distance and the limit obstacle avoidance distance.
[0045] The vehicle active obstacle avoidance control system of the present invention detects and identifies traffic images by adopting an improved YOLOv8 network model, and can obtain the detailed distribution of vehicles to be tested on the lane, avoiding the situation where radar causes incorrect identification of obstacle vehicles, and then screens the vehicles to be tested, removes vehicles to be tested that do not affect the driving safety of the target vehicle, and only retains the obstacle vehicles in front of the same lane, the obstacle vehicles in front of the side lane, and the obstacle vehicles in the rear of the side lane. The three vehicles to be tested are used as obstacles for radar ranging and speed measurement, which effectively reduces the detection range of the laser radar, reduces the amount of calculation, and improves the response efficiency of the vehicle-computer system, making vehicle obstacle avoidance more accurate and safe. By calculating the limit obstacle avoidance distance and the warning obstacle avoidance distance, and then referring to the driving status of the obstacle vehicles in front of the same lane, the obstacle vehicles in front of the side lane, and the obstacle vehicles in the rear of the side lane, the driver is given a risk warning or the target vehicle is actively controlled to slow down, avoiding the situation where the driver causes a collision due to distracted driving, ensuring the safety of the target vehicle in obstacle avoidance, and reducing the collision risk.
[0046] The present invention also provides a readable storage medium having a computer program stored thereon, which implements the above-mentioned vehicle active obstacle avoidance control method when executed by a processor.
[0047] The present invention also proposes a terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned vehicle active obstacle avoidance control method when executing the computer program. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0049] Figure 1 This is a flow chart of the vehicle active obstacle avoidance control method of the present invention;
[0050] Figure 2 This is a structural block diagram of the vehicle active obstacle avoidance control system of the present invention. DETAILED DESCRIPTION
[0051] The following describes embodiments of the present invention in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the embodiments of the present invention, and should not be construed as limiting the present invention.
[0052] In the description of the embodiments of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention.
[0053] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, "plurality" means two or more, unless otherwise specifically defined.
[0054] In the embodiments of the present invention, unless otherwise expressly specified or limited, the terms "installed," "connected," "connected," "fixed," etc. should be understood in a broad sense. For example, they may refer to fixed connection, detachable connection, or integration; mechanical connection or electrical connection; direct connection or indirect connection through an intermediate medium; internal communication between two components or interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in the embodiments of the present invention based on specific circumstances.
[0055] Example 1
[0056] Reference Figure 1 The first embodiment of the present invention is a vehicle active obstacle avoidance control method, the method comprising:
[0057] S1: Obtain a traffic image of the target vehicle and detect and identify the traffic image using the pre-trained improved YOLOv8 network model to obtain multiple vehicles to be tested in the traffic image. Due to the presence of other obstacles such as green belts on both sides of the road, the improved YOLOv8 network model is used to detect and identify the traffic image to ensure accurate identification of vehicles on the road.
[0058] Furthermore, the improved YOLOv8 network model includes a backbone network, a neck network and a detection layer. A GAM attention mechanism module is added before the neck network is connected to the medium-sized detection head and the large-sized detection head of the detection layer. The GSBott1eneck lightweight bottleneck layer is introduced based on the GSConv lightweight convolution. The one-time aggregation method is used to construct the VoV-GSCSP cross-level partial network module, and the VoV-GSCSP cross-level partial network module is used to replace the original C2f module of the neck network. The regression loss function of the improved YOLOv8 network model is the WIoU_v3 loss function.
[0059] In this embodiment, the GAM attention mechanism module introduces a channel attention module of 3D arrangement and multi-layer perception, and also introduces a convolutional spatial attention module; the channel attention module retains feature information through 3D arrangement across three dimensions, and uses a multi-layer perceptron to amplify the cross-dimensional channel and space dependencies, and the convolutional spatial attention module uses two convolutional layers for spatial information fusion, and reduces the negative impact of information on detection results through depooling operations; the neck network introduces GSConv lightweight convolution to reduce the number of model parameters and computational complexity, and maximize the detection effect. The VoV-GSCSP cross-level partial network module can further reduce the number of model parameters and computational complexity.
[0060] Furthermore, the WIoU_v3 loss function is:
[0061] P WIoU_v3 =θ×k WIoU ×p IoU
[0062]
[0063]
[0064] Among them, P WIoU_v3 is the WIoU_v3 loss function, θ is the non-monotonic focusing coefficient, k WIoU is the normalized distance between the center point of the real box and the predicted box, p IoU is the loss value of the bounding box IoU, x is the horizontal coordinate of the center point of the real box, y is the vertical coordinate of the center point of the real box, and x j is the horizontal coordinate of the center point of the prediction box, y j is the vertical coordinate of the center point of the prediction box, w j is the width of the minimum prediction box, h j is the height of the minimum prediction box, ω is the outlier degree, is the gradient gain coefficient, and i is a constant.
[0065] In this embodiment, the WIoU_v3 loss function P WIoU_v3The gradient allocation strategy of the non-monotonic dynamic focusing mechanism allocates gradient gains to high-quality anchor frames in the early stage, and allocates gradient gains to low-quality anchor frames in the middle and late stages to reduce harmful gradients of low-quality images. Therefore, it not only enables the model to obtain better generalization ability, but also improves the regression accuracy of the network bounding box.
[0066] S2: Obtain the position information of the vehicle to be tested, and screen out the front obstacle vehicle in the same lane, the front obstacle vehicle in the side lane, and the rear obstacle vehicle in the side lane from the vehicles to be tested based on the position information; in this embodiment, the target vehicle is traveling on a one-way two-lane road.
[0067] Furthermore, the steps of obtaining the position information of the vehicle to be tested and screening out the vehicle with the front obstacle in the same lane, the vehicle with the front obstacle in the side lane, and the vehicle with the rear obstacle in the side lane from the vehicles to be tested according to the position information specifically include:
[0068] A two-dimensional coordinate system is established on the traffic image with the target vehicle as the origin. Coordinate points are defined for the vehicle under test on the traffic image, with the coordinate points of the vehicle under test serving as the location information. The horizontal axis of the two-dimensional coordinate system is perpendicular to the lane, and the vertical axis is parallel to the lane. The coordinate point of the vehicle under test is the center point of the identification frame of the vehicle under test, with the origin being on the centerline of the current lane.
[0069] According to the lane width and the coordinates of the vehicle to be tested, the vehicle to be tested that belongs to the same lane as the target vehicle is defined as the same lane vehicle, and the vehicle to be tested that does not belong to the same lane as the target vehicle is defined as the side lane vehicle;
[0070] The vehicle with the smallest positive vertical coordinate among the vehicles in the same lane is defined as the front obstacle vehicle in the same lane, the vehicle with the smallest positive vertical coordinate among the vehicles in the side lane is defined as the front obstacle vehicle in the side lane, and the vehicle with the largest vertical coordinate less than or equal to zero among the vehicles in the side lane is defined as the rear obstacle vehicle in the side lane.
[0071] S3: Based on the vehicle-mounted laser radar, the current distances of the target vehicle to the obstacle vehicle in front of the same lane, the obstacle vehicle in front of the side lane, and the obstacle vehicle in the rear of the side lane are obtained, and the current speeds of the obstacle vehicle in front of the same lane and the obstacle vehicle in front of the side lane are obtained. In this embodiment, the technology of the vehicle-mounted laser radar being able to obtain the speed of other vehicles on the road and the distance to the target vehicle is well known to those skilled in the art and will not be described in detail here.
[0072] S4: Obtaining a limit obstacle avoidance distance and a warning obstacle avoidance distance based on the current distance from the target vehicle to the obstacle vehicle in front of the same lane and the current speed of the obstacle vehicle in front of the same lane;
[0073] Furthermore, the maximum obstacle avoidance distance is:
[0074]
[0075] Among them, L e is the maximum obstacle avoidance distance, v a is the current speed of the target vehicle a, v b is the current speed of the obstacle vehicle b in the same lane, t1 is the driver's reaction time, t2 is the response delay of the vehicle computer, μ is the road adhesion coefficient, g is the acceleration of gravity, l ab is the current distance between target vehicle a and the obstacle vehicle b in the same lane when the current distance between the target vehicle and the obstacle vehicle in front of the same lane is less than or equal to the limit obstacle avoidance distance and the target vehicle brakes and decelerates until its speed matches that of the obstacle vehicle in front of the same lane.
[0076] Furthermore, the warning obstacle avoidance distance is:
[0077]
[0078] Among them, t3 is the driver's warning reaction time, t3 = 2s.
[0079] S5: If the current distance between the target vehicle and the obstacle vehicle in front of the same lane is greater than or equal to the warning obstacle avoidance distance, no warning prompt will be issued. If the current distance between the target vehicle and the obstacle vehicle in front of the same lane is less than or equal to the limit obstacle avoidance distance, a collision prompt will be issued and the target vehicle will be controlled to brake and decelerate until the target vehicle and the obstacle vehicle in front of the same lane remain relatively stationary.
[0080] S6: If the current distance between the target vehicle and the vehicle with the same lane obstacle in front is between the warning obstacle avoidance distance and the limit obstacle avoidance distance, an obstacle avoidance warning is issued. At the same time, a decision is made whether to issue a lane change prompt based on the current speeds of the vehicle with the same lane obstacle in front and the vehicle with the side lane obstacle in front, as well as the current distances between the target vehicle and the vehicle with the same lane obstacle in front, the vehicle with the side lane obstacle in front, and the vehicle with the side lane obstacle in back, respectively.
[0081] Furthermore, if the current distance between the target vehicle and the vehicle with the same lane obstacle in front is between the warning obstacle avoidance distance and the limit obstacle avoidance distance, an obstacle avoidance warning is issued, and at the same time, based on the current speeds of the vehicle with the same lane obstacle in front and the vehicle with the side lane obstacle in front, and the current distances of the target vehicle to the vehicle with the same lane obstacle in front, the vehicle with the side lane obstacle in front, and the vehicle with the side lane obstacle in back, respectively, the steps of determining whether to issue a lane change prompt specifically include:
[0082] If the current distance between the target vehicle and the obstacle vehicle ahead in the same lane is between the warning obstacle avoidance distance and the limit obstacle avoidance distance, an obstacle avoidance warning is issued; and if either the first constraint or the second constraint is met at the same time, the vehicle computer issues a lane change prompt;
[0083] The first constraint is:
[0084] v a >v c >vb
[0085] (L ac >L f )U(L ad >L e )
[0086] The second constraint is:
[0087] v b <v a <v c
[0088] (L ac >L e )U(L ad >L e )
[0089] Among them, v c is the current speed of the obstacle vehicle c in the side lane, L ac is the current distance from the target vehicle a to the obstacle vehicle c in the front lane, L ad is the current distance from the target vehicle a to the obstacle vehicle d on the rear side of the side road.
[0090] The vehicle active obstacle avoidance control method of the present invention detects and identifies traffic images by adopting an improved YOL0v8 network model, and can obtain the detailed distribution of vehicles to be tested on the lane, avoid the situation where radar causes incorrect identification of obstacle vehicles, and then screen the vehicles to be tested, remove the vehicles to be tested that do not affect the driving safety of the target vehicle, and only retain the obstacle vehicles in the front of the same lane, the obstacle vehicles in the front of the side lane, and the obstacle vehicles in the rear of the side lane. The three vehicles to be tested are used as obstacles for radar ranging and speed measurement, which effectively reduces the detection range of the laser radar, reduces the amount of calculation, and improves the response efficiency of the vehicle-computer system, making vehicle obstacle avoidance more accurate and safe; by calculating the limit obstacle avoidance distance and the warning obstacle avoidance distance, and then referring to the driving status of the obstacle vehicles in the front of the same lane, the obstacle vehicles in the front of the side lane, and the obstacle vehicles in the rear of the side lane, the driver is given a risk warning or the target vehicle is actively controlled to slow down, so as to avoid the driver from colliding due to distracted driving, ensure the safety of the target vehicle in obstacle avoidance, and reduce the collision risk.
[0091] Example 2
[0092] Reference Figure 2 The present invention also proposes a vehicle active obstacle avoidance control system, the system comprising:
[0093] The first acquisition module is used to acquire a traffic image in front of the target vehicle, detect and identify the traffic image based on the pre-trained improved YOLOv8 network model, and obtain multiple vehicles to be tested in the traffic image;
[0094] A screening module is used to obtain the position information of the vehicle to be tested and screen out the vehicles with obstacles in front of the same lane, the vehicles with obstacles in front of the side lane and the vehicles with obstacles in the rear of the side lane from the vehicles to be tested based on the position information;
[0095] The second acquisition module is used to obtain the current distance of the target vehicle to the obstacle vehicle in front of the same lane, the obstacle vehicle in front of the side lane, and the obstacle vehicle in the rear of the side lane based on the vehicle-mounted laser radar, and simultaneously obtain the current speed of the obstacle vehicle in front of the same lane and the obstacle vehicle in front of the side lane;
[0096] A calculation module is used to obtain a limit obstacle avoidance distance and a warning obstacle avoidance distance based on the current distance from the target vehicle to the obstacle vehicle in front of the same lane and the current speed of the obstacle vehicle in front of the same lane;
[0097] The first judgment module is used to determine whether the target vehicle maintains its driving state if the current distance from the target vehicle to the obstacle vehicle in front of it is greater than or equal to the warning obstacle avoidance distance; and whether the current distance from the target vehicle to the obstacle vehicle in front of it is less than or equal to the limit obstacle avoidance distance, and whether the target vehicle brakes and decelerates;
[0098] The second judgment module is used to determine whether the target vehicle should change lanes if the current distance from the target vehicle to the obstacle vehicle in front of the same lane is between the warning obstacle avoidance distance and the limit obstacle avoidance distance, based on the current speeds of the obstacle vehicles in front of the same lane and the obstacle vehicles in front of the side lane, as well as the current distances from the target vehicle to the obstacle vehicles in front of the same lane, the obstacle vehicles in front of the side lane, and the obstacle vehicles in the rear of the side lane, respectively.
[0099] Furthermore, the improved YOLOv8 network model includes a backbone network, a neck network and a detection layer. A GAM attention mechanism module is added before the neck network is connected to the medium-sized detection head and the large-sized detection head of the detection layer. The GSBottleneck lightweight bottleneck layer is introduced based on the GSConv lightweight convolution. The one-time aggregation method is used to construct the VoV-GSCSP cross-level partial network module, and the VoV-GSCSP cross-level partial network module is used to replace the original C2f module of the neck network. The regression loss function of the improved YOLOv8 network model is the WIoU_v3 loss function.
[0100] Furthermore, the WIoU_v3 loss function is:
[0101] P WIoU_v3 =θ×k WIoU ×p IoU
[0102]
[0103]
[0104] Among them, P WIoU_v3is the WIoU_v3 loss function, θ is the non-monotonic focusing coefficient, k WIoU is the normalized distance between the center point of the real box and the predicted box, p IoU is the loss value of the bounding box IoU, x is the horizontal coordinate of the center point of the real box, y is the vertical coordinate of the center point of the real box, and x j is the horizontal coordinate of the center point of the prediction box, y j is the vertical coordinate of the center point of the prediction box, w j is the width of the minimum prediction box, h j is the height of the minimum prediction box, ω is the outlier degree, is the gradient gain coefficient, and i is a constant.
[0105] Furthermore, the screening module is specifically used to:
[0106] A two-dimensional coordinate system is established on the traffic image with the target vehicle as the origin. Coordinate points are defined for the vehicle under test on the traffic image, with the coordinate points of the vehicle under test serving as the location information. The horizontal axis of the two-dimensional coordinate system is perpendicular to the lane, and the vertical axis is parallel to the lane. The coordinate point of the vehicle under test is the center point of the identification frame of the vehicle under test, with the origin being on the centerline of the current lane.
[0107] According to the lane width and the coordinates of the vehicle to be tested, the vehicle to be tested that belongs to the same lane as the target vehicle is defined as the same lane vehicle, and the vehicle to be tested that does not belong to the same lane as the target vehicle is defined as the side lane vehicle;
[0108] The vehicle with the smallest positive vertical coordinate among the vehicles in the same lane is defined as the front obstacle vehicle in the same lane, the vehicle with the smallest positive vertical coordinate among the vehicles in the side lane is defined as the front obstacle vehicle in the side lane, and the vehicle with the largest vertical coordinate less than or equal to zero among the vehicles in the side lane is defined as the rear obstacle vehicle in the side lane.
[0109] Furthermore, the maximum obstacle avoidance distance is:
[0110]
[0111] Among them, L e is the maximum obstacle avoidance distance, v a is the current speed of the target vehicle a, v b is the current speed of the obstacle vehicle b in the same lane, t1 is the driver's reaction time, t2 is the response delay of the vehicle computer, μ is the road adhesion coefficient, g is the acceleration of gravity, l ab is the current distance between target vehicle a and the obstacle vehicle b in the same lane when the current distance between the target vehicle and the obstacle vehicle in front of the same lane is less than or equal to the limit obstacle avoidance distance and the target vehicle brakes and decelerates until its speed matches that of the obstacle vehicle in front of the same lane.
[0112] Furthermore, the warning obstacle avoidance distance is:
[0113]
[0114] Among them, t3 is the driver's warning reaction time, t3 = 2s.
[0115] Furthermore, the second judgment module is specifically configured to:
[0116] If the current distance between the target vehicle and the obstacle vehicle ahead in the same lane is between the warning obstacle avoidance distance and the limit obstacle avoidance distance, an obstacle avoidance warning is issued; and if either the first constraint or the second constraint is met at the same time, the vehicle computer issues a lane change prompt;
[0117] The first constraint is:
[0118] v a >v c >v b
[0119] (L ac >L f )U(L ad >L e )
[0120] The second constraint is:
[0121] v b <v a <v c
[0122] (L ac >L e )U(L ad >L e )
[0123] Among them, v c is the current speed of the obstacle vehicle c in the side lane, L ac is the current distance from the target vehicle a to the obstacle vehicle c in the front lane, L ad is the current distance from the target vehicle a to the obstacle vehicle d on the rear side of the side road.
[0124] The vehicle active obstacle avoidance control system of the present invention detects and identifies traffic images by adopting an improved YOLOv8 network model, and can obtain the detailed distribution of vehicles to be tested on the lane, avoiding the situation where radar causes incorrect identification of obstacle vehicles, and then screens the vehicles to be tested, removes vehicles to be tested that do not affect the driving safety of the target vehicle, and only retains the obstacle vehicles in front of the same lane, the obstacle vehicles in front of the side lane, and the obstacle vehicles in the rear of the side lane. The three vehicles to be tested are used as obstacles for radar ranging and speed measurement, which effectively reduces the detection range of the laser radar, reduces the amount of calculation, and improves the response efficiency of the vehicle-computer system, making vehicle obstacle avoidance more accurate and safe. By calculating the limit obstacle avoidance distance and the warning obstacle avoidance distance, and then referring to the driving status of the obstacle vehicles in front of the same lane, the obstacle vehicles in front of the side lane, and the obstacle vehicles in the rear of the side lane, the driver is given a risk warning or the target vehicle is actively controlled to slow down, avoiding the situation where the driver causes a collision due to distracted driving, ensuring the safety of the target vehicle in obstacle avoidance, and reducing the collision risk.
[0125] Example 3
[0126] The third embodiment of the present invention is based on the same inventive concept. The present invention proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the vehicle active obstacle avoidance control method of the above embodiment.
[0127] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, "computer-readable medium" can be any device that stores, communicates, propagates, or transmits a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.
[0128] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.
[0129] Among them, the memory may include a large-capacity memory for data or instructions. By way of example and not limitation, the memory may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory may include a removable or non-removable (or fixed) medium. Where appropriate, the memory may be inside or outside the data processing device. In a specific embodiment, the memory is a non-volatile memory. In a specific embodiment, the memory includes a read-only memory (ROM) and a random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM) or a flash memory (FLASH), or a combination of two or more of these. Under appropriate circumstances, the RAM can be a static random access memory (SRAM) or a dynamic random access memory (DRAM), where the DRAM can be a fast page mode dynamic random access memory (FPMDRAM), an extended data output dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.
[0130] Example 4
[0131] The fourth embodiment of the present invention is based on the same inventive concept. The present invention proposes a terminal, which includes: a processor and a memory; the processor and the memory communicate with each other; the memory is used to store instructions; the processor is used to execute the instructions in the memory to execute the vehicle active obstacle avoidance control method of the above embodiment.
[0132] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0133] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
[0134] Under the premise that no conflict occurs, those skilled in the art may freely combine and superimpose the above-mentioned additional technical features.
[0135] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.
Claims
1. A vehicle active obstacle avoidance control method, characterized in that: The method comprises: Acquire a traffic image of a target vehicle, detect and identify the traffic image according to a pre-trained improved YOLOv8 network model, and obtain multiple vehicles to be tested in the traffic image; Acquire the position information of the vehicle to be tested, and select the vehicle with a front obstacle on the same lane, the vehicle with a front obstacle on the side lane, and the vehicle with a rear obstacle on the side lane from the vehicles to be tested according to the position information; Based on the vehicle-mounted laser radar, the current distances of the target vehicle to the obstacle vehicle in front of the same lane, the obstacle vehicle in front of the side lane, and the obstacle vehicle in the rear of the side lane are obtained, and the current speeds of the obstacle vehicle in front of the same lane and the obstacle vehicle in front of the side lane are obtained; Obtaining a limit obstacle avoidance distance and a warning obstacle avoidance distance according to a current distance from the target vehicle to the obstacle vehicle in front of the same lane and a current speed of the obstacle vehicle in front of the same lane; If the current distance from the target vehicle to the obstacle vehicle in front of the same lane is greater than or equal to the warning obstacle avoidance distance, no warning prompt is given; if the current distance from the target vehicle to the obstacle vehicle in front of the same lane is less than or equal to the limit obstacle avoidance distance, a collision prompt is given and the target vehicle is controlled to brake and decelerate; If the current distance from the target vehicle to the vehicle with the front obstacle in the same lane is between the warning obstacle avoidance distance and the limit obstacle avoidance distance, an obstacle avoidance warning is issued. At the same time, according to the current speeds of the vehicle with the front obstacle in the same lane and the vehicle with the front obstacle in the side lane, and the current distances from the target vehicle to the vehicle with the front obstacle in the same lane, the vehicle with the front obstacle in the side lane, and the vehicle with the rear obstacle in the side lane, it is determined whether to issue a lane change prompt.
2. The vehicle active obstacle avoidance control method according to claim 1, characterized in that: The improved YOLOv8 network model includes a backbone network, a neck network and a detection layer. A GAM attention mechanism module is added before the neck network is connected with the medium-sized detection head and the large-sized detection head of the detection layer. A GSbottleneck lightweight bottleneck layer is introduced based on GSConv lightweight convolution. A VoV-GSCSP cross-level partial network module is constructed using a one-time aggregation method. The VoV-GSCSP cross-level partial network module is used to replace the original C2f module of the neck network. The regression loss function of the improved YOLOv8 network model is a WIoU_v3 loss function.
3. The vehicle active obstacle avoidance control method according to claim 2, characterized in that: The WIoU_v3 loss function is: P WIoU_v3 =θ×k WIoU ×p IoU Among them, P WIoU_v3 is the WIoU_v3 loss function, θ is the non-monotonic focusing coefficient, k WIoU is the normalized distance between the center point of the real box and the predicted box, p IoU is the loss value of the bounding box IoU, x is the horizontal coordinate of the center point of the real box, y is the vertical coordinate of the center point of the real box, and x j is the horizontal coordinate of the center point of the prediction box, y j is the vertical coordinate of the center point of the prediction box, w j is the width of the minimum prediction box, h j is the height of the minimum prediction box, ω is the outlier degree, is the gradient gain coefficient, and i is a constant.
4. The vehicle active obstacle avoidance control method according to claim 1, characterized in that: The step of obtaining the position information of the vehicle to be tested, and screening out the front obstacle vehicle on the same lane, the front obstacle vehicle on the side lane, and the rear obstacle vehicle on the side lane from the vehicles to be tested according to the position information specifically includes: Taking the target vehicle as the origin, a two-dimensional coordinate system is established on the traffic image, and coordinate points are defined for the vehicle to be tested on the traffic image, and the coordinate points of the vehicle to be tested are used as position information; wherein the horizontal axis of the two-dimensional coordinate system is perpendicular to the lane, the vertical axis is parallel to the lane, the coordinate point of the vehicle to be tested is the center point of the identification frame of the vehicle to be tested, and the origin is on the center line of the current lane; According to the width of the lane and the coordinate point of the vehicle to be tested, the vehicle to be tested that belongs to the same lane as the target vehicle is defined as a same-lane vehicle, and the vehicle to be tested that does not belong to the same lane as the target vehicle is defined as a side-lane vehicle; The vehicle to be tested whose longitudinal coordinate is the smallest positive value among the vehicles in the same lane is defined as a front obstacle vehicle in the same lane, the vehicle to be tested whose longitudinal coordinate is the smallest positive value among the vehicles in the side lane is defined as a front obstacle vehicle in the side lane, and the vehicle to be tested whose longitudinal coordinate is less than or equal to zero and has the largest value among the vehicles in the side lane is defined as a rear obstacle vehicle in the side lane.
5. The vehicle active obstacle avoidance control method according to claim 1, characterized in that: The limit obstacle avoidance distance is: Among them, L e is the limit obstacle avoidance distance, v a is the current speed of the target vehicle a, v b is the current speed of the obstacle vehicle b in the same lane, t1 is the driver's reaction time, t2 is the response delay of the vehicle computer, μ is the road adhesion coefficient, g is the gravitational acceleration, l ab is the current distance between the target vehicle a and the front obstacle vehicle b in the same lane when the current distance between the target vehicle and the front obstacle vehicle in the same lane is less than or equal to the limit obstacle avoidance distance and the target vehicle brakes and decelerates until its speed is consistent with that of the front obstacle vehicle in the same lane.
6. The vehicle active obstacle avoidance control method according to claim 5, characterized in that: The warning obstacle avoidance distance is: Wherein, t3 is the driver's warning reaction time, t3 = 2s.
7. The vehicle active obstacle avoidance control method according to claim 6, characterized in that: If the current distance between the target vehicle and the front obstacle vehicle in the same lane is between the warning obstacle avoidance distance and the limit obstacle avoidance distance, an obstacle avoidance warning is issued, and at the same time, according to the current vehicle speeds of the front obstacle vehicle in the same lane and the front obstacle vehicle in the side lane, and the current distances between the target vehicle and the front obstacle vehicle in the same lane, the front obstacle vehicle in the side lane, and the rear obstacle vehicle in the side lane, the steps of determining whether to issue a lane change prompt include: If the current distance between the target vehicle and the obstacle vehicle in front of the same lane is between the warning obstacle avoidance distance and the limit obstacle avoidance distance, an obstacle avoidance warning is issued; and if any one of the first constraint condition or the second constraint condition is met at the same time, the vehicle computer issues a lane change prompt; The first constraint condition is: v a >v c >v b (L ac >L f )U(L ad >L e ) The second constraint is: in b <v a <v c (L ac >L e )U(L ad >L e ) Among them, v c is the current speed of the obstacle vehicle c in front of the side road, L ac is the current distance from the target vehicle a to the obstacle vehicle c in front of the side road, L ad is the current distance from the target vehicle a to the obstacle vehicle d at the rear side of the side road.
8. The vehicle active obstacle avoidance control system is characterized by: The system comprises: A first acquisition module is used to acquire a traffic image of a target vehicle, detect and identify the traffic image according to a pre-trained improved YOLOv8 network model, and obtain multiple vehicles to be tested in the traffic image; A screening module, used for obtaining the position information of the vehicle to be tested, and screening out the vehicle with a front obstacle on the same lane, the vehicle with a front obstacle on the side lane, and the vehicle with a rear obstacle on the side lane from the vehicles to be tested according to the position information; The second acquisition module is used to acquire the current distances of the target vehicle to the front obstacle vehicle on the same lane, the front obstacle vehicle on the side lane, and the rear obstacle vehicle on the side lane based on the vehicle-mounted laser radar, and simultaneously acquire the current vehicle speeds of the front obstacle vehicle on the same lane and the front obstacle vehicle on the side lane; A calculation module, used for obtaining a limit obstacle avoidance distance and a warning obstacle avoidance distance according to a current distance from the target vehicle to the front obstacle vehicle on the same road and a current speed of the front obstacle vehicle on the same road; A first judgment module is used for not giving a warning prompt if the current distance between the target vehicle and the obstacle vehicle in front of the same lane is greater than or equal to the warning obstacle avoidance distance, and for giving a collision prompt and controlling the target vehicle to brake and decelerate if the current distance between the target vehicle and the obstacle vehicle in front of the same lane is less than or equal to the limit obstacle avoidance distance; The second judgment module is used to issue an obstacle avoidance warning if the current distance from the target vehicle to the front obstacle vehicle in the same lane is between the warning obstacle avoidance distance and the limit obstacle avoidance distance, and at the same time determine whether to issue a lane change prompt according to the current vehicle speeds of the front obstacle vehicle in the same lane and the front obstacle vehicle in the side lane, and the current distances from the target vehicle to the front obstacle vehicle in the same lane, the front obstacle vehicle in the side lane, and the rear obstacle vehicle in the side lane, respectively.
9. A readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the vehicle active obstacle avoidance control method as described in any one of claims 1 to 7 is implemented.
10. A terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the vehicle active obstacle avoidance control method as described in any one of claims 1 to 7 is implemented.
