Obstacle avoidance method and device of vehicle and storage medium

Abstract

The application discloses an obstacle avoidance method and device of a vehicle and a storage medium, and belongs to the technical field of vehicle control. The method comprises the following steps: collecting an image and a radar detection result of a driving direction of a road where the vehicle is located; detecting the image of the driving direction of the road where the vehicle is located through an obstacle identification model to obtain an image detection result; in response to at least one of the radar detection result indicating that there is an obstacle within a first distance in front of the vehicle or the image detection result indicating that there is an obstacle within a second distance in front of the vehicle, determining whether the obstacle needs to be avoided based on a type of the obstacle and a feature of the obstacle; in response to the obstacle needing to be avoided, predicting a future motion trajectory of the obstacle; determining whether the vehicle needs to be adjusted to avoid the obstacle based on the future motion trajectory of the obstacle; and in response to the vehicle needing to be adjusted to avoid the obstacle, adjusting a driving state of the vehicle based on the future motion trajectory of the obstacle. The flexibility of the vehicle obstacle avoidance control is improved, and the reliability and safety of automatic driving are improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of vehicle control, in particular to an obstacle avoidance method and device for a vehicle and a storage medium. BACKGROUND

[0002] With the development of intelligent driving technology, in order to ensure driving safety and improve driving convenience, during vehicle driving, the vehicle needs to be controlled to avoid obstacles in the road and avoid obstacles in the road. In related technologies, obstacles in the road are monitored, and if obstacles such as roadblocks, road signs or pedestrians are detected, the occupants in the vehicle are prompted.

[0003] In related technologies, it is easy to miss irregular obstacles in the road, or there may be inaccurate identification of the type of obstacle, which will affect the vehicle's immediate response and endanger driving safety. Therefore, how to monitor irregular obstacles in the road where the vehicle is located, and when irregular obstacles are detected, control the vehicle to make corresponding adjustments according to the type and characteristics of the irregular obstacles, is very important to improve the reliability and safety of autonomous driving. SUMMARY

[0004] Embodiments of the present application provide an obstacle avoidance method and device for a vehicle and a storage medium, which can be used to improve the reliability and safety of autonomous driving. The technical solution is as follows:

[0005] On the one hand, the present application provides an obstacle avoidance method for a vehicle, the method comprising:

[0006] Collecting images and radar detection results in the driving direction of the road where the vehicle is located, the radar detection results being used to indicate whether there is an obstacle within a first distance in front of the vehicle;

[0007] Detecting the images in the driving direction of the road where the vehicle is located through an obstacle recognition model to obtain image detection results, the image detection results being used to indicate whether there is the obstacle within a second distance in front of the vehicle, the type of the obstacle and the characteristics of the obstacle;

[0008] In response to at least one of the radar detection results indicating that there is the obstacle within the first distance in front of the vehicle or the image detection results indicating that there is the obstacle within the second distance in front of the vehicle, determining a first detection result based on the type of the obstacle and the characteristics of the obstacle, the first detection result being used to indicate whether the obstacle needs to be avoided;

[0009] In response to the first detection result indicating that the obstacle needs to be avoided, predicting a future motion trajectory of the obstacle;

[0010] determine a second detection result based on the future motion trajectory of the obstacle, the second detection result being used to indicate whether the vehicle needs to adjust to avoid the obstacle;

[0011] adjust a driving state of the vehicle based on the future motion trajectory of the obstacle in response to the second detection result indicating that the vehicle needs to adjust to avoid the obstacle.

[0012] In another aspect, an obstacle avoidance device for a vehicle is provided, the device comprising:

[0013] an acquisition module configured to acquire an image of a driving direction of a road where the vehicle is located and a radar detection result, the radar detection result being used to indicate whether there is an obstacle within a first distance in front of the vehicle;

[0014] a detection module configured to detect the image of the driving direction of the road where the vehicle is located by an obstacle recognition model to obtain an image detection result, the image detection result being used to indicate whether there is the obstacle within a second distance in front of the vehicle, a type of the obstacle, and a feature of the obstacle;

[0015] a first determination module configured to determine a first detection result based on the type of the obstacle and the feature of the obstacle in response to at least one of the radar detection result indicating that there is the obstacle within the first distance in front of the vehicle or the image detection result indicating that there is the obstacle within the second distance in front of the vehicle, the first detection result being used to indicate whether the obstacle needs to be avoided;

[0016] a prediction module configured to predict a future motion trajectory of the obstacle in response to the first detection result indicating that the obstacle needs to be avoided;

[0017] a second determination module configured to determine a second detection result based on the future motion trajectory of the obstacle, the second detection result being used to indicate whether the vehicle needs to adjust to avoid the obstacle;

[0018] an adjustment module configured to adjust a driving state of the vehicle based on the future motion trajectory of the obstacle in response to the second detection result indicating that the vehicle needs to adjust to avoid the obstacle.

[0019] In another aspect, a non-transitory computer-readable storage medium is also provided, the computer-readable storage medium storing at least one computer program, the at least one computer program being loaded and executed by a processor to enable a computer to implement the obstacle avoidance method for a vehicle as described above.

[0020] In another aspect, a computer program product or computer program is also provided, which includes computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to cause the computer device to perform the obstacle avoidance method of any of the above aspects.

[0021] The technical solutions provided in the present application at least have the following beneficial effects:

[0022] The present application determines whether there is an obstacle in front of the vehicle through image detection results and radar detection results, determines through image detection results and radar detection results, avoids the situation of missing detection due to the special shape of the special-shaped obstacle, and improves the accuracy of obstacle detection; in response to at least one of the radar detection result being an obstacle or the image detection result being an obstacle, it is determined whether the obstacle needs to be avoided based on the type and characteristics of the obstacle, so that the obstacle is avoided only when the characteristics of the obstacle meet the requirements of the type of obstacle that needs to be avoided, and the accuracy of vehicle obstacle avoidance is improved. After confirming that the obstacle needs to be avoided, the future motion trajectory of the obstacle is predicted, and the driving state of the vehicle is adjusted based on the future motion trajectory of the obstacle to avoid the obstacle, improving the flexibility of vehicle obstacle avoidance control, thereby improving the reliability and safety of autonomous driving. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0024] Figure 1 is a schematic diagram of an implementation environment provided by an embodiment of the present application;

[0025] Figure 2 is a flowchart of an obstacle avoidance method of a vehicle provided by an embodiment of the present application;

[0026] Figure 3 is a structural schematic diagram of an obstacle avoidance device of a vehicle provided by an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of the present application more clear, the embodiments of the present application will be further described in detail below with reference to the drawings.

[0028] An obstacle avoidance method of a vehicle is provided in the embodiments of the present application, please refer to Figure 1As shown in the schematic diagram, the method implementation environment provided by the embodiment of the present application includes a navigation system 11, a radar device 12, a camera 13, an autonomous driving domain controller 14, a central control large screen 15, a steering system 16, and a power system 17.

[0029] Optionally, the autonomous driving domain controller 14 acquires an image of a driving direction of a road where the vehicle is located collected by the camera 13, and acquires a radar detection result of the driving direction of the road where the vehicle is located collected by the radar device 12. The autonomous driving domain controller 14 detects the image through a self-provided obstacle recognition model to obtain an image detection result. In response to at least one of the radar detection result indicating that there is an obstacle within a first distance in front of the vehicle or the image detection result indicating that there is an obstacle within a second distance in front of the vehicle, the autonomous driving domain controller 14 determines whether the obstacle needs to be avoided based on a type of the obstacle and a feature of the obstacle. In response to the obstacle needing to be avoided, the autonomous driving domain controller 14 controls the central control large screen 15 of the vehicle to display a text prompt that there is an obstacle in front of the vehicle, a type of the obstacle, and a distance between the vehicle and the obstacle. The autonomous driving domain controller 14 predicts a future motion trajectory of the obstacle, and acquires a traffic condition around the road where the vehicle is located from the navigation system 11. The autonomous driving domain controller 14 further controls the vehicle to change lanes through the steering system 16 or controls the vehicle to accelerate or decelerate through the power system 17 according to the traffic condition around the road where the vehicle is located and the future motion trajectory of the obstacle, so as to complete obstacle avoidance of the vehicle.

[0030] The navigation system 11, the radar device 12, the camera 13, the autonomous driving domain controller 14, the central control large screen 15, the steering system 16, and the power system 17 are connected through wired or wireless communication.

[0031] Based on the above Figure 1 The embodiment of the present application provides a vehicle obstacle avoidance method as shown in the schematic diagram. Figure 2 The method includes steps 201-206.

[0032] In step 201, an image of a driving direction of a road where the vehicle is located and a radar detection result are collected, and the radar detection result is used to indicate whether there is an obstacle within a first distance in front of the vehicle.

[0033] In a possible implementation, the camera and the radar device are installed on the vehicle, wherein the camera and the radar device can be installed at the front of the vehicle, and the radar device can include at least one of a laser radar and a millimeter wave radar. Illustratively, the image and the radar detection result in the driving direction of the road where the vehicle is located are collected, including: the automatic driving domain controller collects the image in the driving direction of the road where the vehicle is located through the camera installed on the vehicle, and collects the radar detection result in the driving direction of the road where the vehicle is located through the radar device installed on the vehicle, wherein the radar detection result is used to indicate whether there is an obstacle within a first distance in front of the vehicle. Optionally, the first distance can be set as the distance of the obstacle that needs to be warned according to requirements.

[0034] Illustratively, the radar detection result in the driving direction of the road where the vehicle is located is collected through the radar device installed on the vehicle, including: the radar installed on the vehicle emits a detection signal to the driving direction of the road where the vehicle is located, if the returned detection signal is received, the distance between the vehicle and the obstacle is calculated based on the time length between the emission of the detection signal and the reception of the returned detection signal corresponding to the detection signal. The calculated distance between the vehicle and the obstacle is compared with the first distance, if the distance between the vehicle and the obstacle is less than or equal to the first distance, the radar detection result indicates that there is an obstacle within the first distance in front of the vehicle. If the returned detection signal corresponding to the detection signal is not received or the distance between the vehicle and the obstacle is greater than the first distance, the radar detection result indicates that there is no obstacle within the first distance in front of the vehicle. In a possible implementation, the formula for calculating the distance between the vehicle and the obstacle based on the time length between the emission of the detection signal and the reception of the returned detection signal corresponding to the detection signal is as follows:

[0035] L = (T x light speed) ÷ 2

[0036] wherein L is the distance between the vehicle and the obstacle, T is the time length between the emission of the detection signal and the reception of the returned detection signal corresponding to the detection signal, the light speed can be 300,000 kilometers per second, the unit of the distance between the vehicle and the obstacle can be kilometers, and the unit of the time length between the emission of the detection signal and the reception of the returned detection signal corresponding to the detection signal can be seconds.

[0037] In step 202, the image is detected through the obstacle recognition model to obtain an image detection result, and the image detection result is used to indicate whether there is an obstacle within a second distance in front of the vehicle, the type of the obstacle and the characteristics of the obstacle.

[0038] In a possible implementation, before detecting the image by the obstacle recognition model, the obstacle recognition model is established, an input of the obstacle recognition model is the image, and an output of the obstacle recognition model is an image detection result, where the image detection result is used to indicate whether there is an obstacle within a second distance in front of the vehicle. If the image detection result indicates that there is an obstacle within the second distance in front of the vehicle, the output of the obstacle recognition model further includes a type of the obstacle and a feature of the obstacle, and the feature of the obstacle includes a size of the obstacle and a shape of the obstacle.

[0039] Illustratively, the obstacle recognition model is established, including: obtaining an obstacle training image and a corresponding image annotation, the image annotation including a type of a first obstacle, a size of the first obstacle, and a shape of the first obstacle; identifying, by an initial obstacle recognition model, a type of a second obstacle, a size of the second obstacle, and a shape of the second obstacle on the obstacle training image; determining a loss value based on the type of the first obstacle and the type of the second obstacle, the size of the first obstacle and the size of the second obstacle, and the shape of the first obstacle and the shape of the second obstacle; adjusting the initial obstacle recognition model based on the loss value to obtain an adjusted initial obstacle recognition model; and in response to the loss value being less than a loss value threshold, taking the adjusted initial obstacle recognition model as the obstacle recognition model.

[0040] Optionally, the obstacle training image is an obstacle image with the type of the first obstacle, the size of the first obstacle, and the shape of the first obstacle being annotated. The initial obstacle recognition model is constructed by selecting a model architecture, for example, a LeNet (convolutional neural network) or AlexNet (deep convolutional neural network) model architecture. Then the obstacle training image is input into the initial obstacle recognition model to output the type of the second obstacle, the size of the second obstacle, and the shape of the second obstacle. The difference between the type of the first obstacle and the type of the second obstacle, the size of the first obstacle and the size of the second obstacle, and the shape of the first obstacle and the shape of the second obstacle is taken as the loss value, and the architecture of the initial obstacle recognition model is adjusted based on the loss value to obtain the adjusted initial obstacle recognition model. If the loss value is greater than or equal to the loss value threshold, the adjusted initial obstacle recognition model is repeatedly trained by using the remaining obstacle training images until the loss value is less than the loss value threshold, and the adjusted initial obstacle recognition model is taken as the obstacle recognition model.

[0041] Illustratively, after the establishment of the obstacle recognition model is completed, an image of a driving direction of a road where the vehicle is located is input into the obstacle recognition model, and an image detection result output by the obstacle recognition model is obtained, where the image detection result is used to indicate whether there is an obstacle within a second distance in front of the vehicle. Optionally, the second distance is a distance that can be captured by a camera.

[0042] In step 203, in response to at least one of the radar detection result indicating that there is an obstacle within the first distance in front of the vehicle or the image detection result indicating that there is an obstacle within the second distance in front of the vehicle, a first detection result is determined based on the type of the obstacle and the characteristics of the obstacle, the first detection result being used to indicate whether the obstacle needs to be avoided.

[0043] Exemplarily, if at least one of the radar detection result indicating that there is an obstacle within the first distance in front of the vehicle or the image detection result indicating that there is an obstacle within the second distance in front of the vehicle is satisfied, a first detection result is determined based on the type of the obstacle and the characteristics of the obstacle, wherein the first detection result is used to indicate whether the obstacle needs to be avoided. In a possible implementation, determining the first detection result based on the type of the obstacle and the characteristics of the obstacle includes: in response to the type of the obstacle being a rockfall and the size of the obstacle being greater than a first threshold value, or the type of the obstacle being plastic and the size of the obstacle being greater than a second threshold value, determining the first detection result to be indicative of the obstacle needing to be avoided.

[0044] Optionally, if the type of the obstacle is a rockfall, the size of the obstacle is compared with the first threshold value, for example, the first threshold value can be the minimum volume of the rockfall that needs to be avoided by the vehicle. The volume of the obstacle is compared with the volume corresponding to the first threshold value, if the volume of the obstacle is greater than the volume corresponding to the first threshold value, it is determined that the first detection result is indicative of the obstacle needing to be avoided. If the volume of the obstacle is less than or equal to the volume corresponding to the first threshold value, it is determined that the first detection result is indicative of the obstacle not needing to be avoided.

[0045] In a possible implementation, if the type of the obstacle is plastic, the size of the obstacle is compared with the second threshold value, for example, the second threshold value can be the minimum area of the plastic that needs to be avoided by the vehicle. The area of the largest face of the obstacle is compared with the area corresponding to the second threshold value, if the area of the largest face of the obstacle is greater than the area corresponding to the second threshold value, it is determined that the first detection result is indicative of the obstacle needing to be avoided. If the area of the largest face of the obstacle is less than or equal to the area corresponding to the second threshold value, it is determined that the first detection result is indicative of the obstacle not needing to be avoided.

[0046] Exemplarily, in response to the first detection result indicating that the obstacle needs to be avoided, the driver of the vehicle is prompted that there is an obstacle in front of the vehicle. Optionally, the prompting manner includes but is not limited to: displaying a text prompt of the existence of the obstacle in front of the vehicle, the type of the obstacle and the distance between the vehicle and the obstacle on a large screen in the vehicle.

[0047] By setting the first threshold value as the minimum volume of the rockfall that the vehicle needs to avoid, and setting the second threshold value as the minimum area of the plastic that the vehicle needs to avoid, different judgment methods are used to judge whether to avoid for different obstacle types, the rockfall is judged by volume, and the plastic is judged by the maximum area of the surface, which is more in line with the characteristics of the obstacle, and improves the accuracy of the vehicle obstacle avoidance.

[0048] In step 204, in response to the first detection result indicating that the obstacle needs to be avoided, the future motion trajectory of the obstacle is predicted.

[0049] Optionally, when it is determined that the first detection result indicates that the obstacle needs to be avoided, the future motion trajectory of the obstacle is predicted, including: target tracking is performed on the obstacle to obtain a motion state of the obstacle; the motion state is analyzed to determine an existing motion trajectory and a current speed of the obstacle; and the future motion trajectory of the obstacle is predicted based on the existing motion trajectory and the current speed of the obstacle.

[0050] In a possible implementation, the target tracking is performed on the obstacle by Kalman filtering to obtain the motion state of the obstacle, and the motion state is analyzed to determine the existing motion trajectory and the current speed of the obstacle. A kinematic model is selected to predict the next-stage motion trajectory and speed of the obstacle, and the predicted next-stage motion trajectory and speed are corrected according to the next-stage motion trajectory and speed actually tracked and identified, to obtain a corrected kinematic model, and the above steps are repeated to complete the prediction of the future motion trajectory of the obstacle. Optionally, the motion model includes but is not limited to: a uniform speed model, a uniform acceleration model, a constant angular velocity model, etc., which can be selected according to experience.

[0051] By predicting the future motion trajectory of the obstacle, it is convenient to subsequently judge whether to adjust the vehicle to avoid the obstacle and how to adjust the driving state of the vehicle.

[0052] In step 205, a second detection result is determined based on the future motion trajectory of the obstacle, and the second detection result is used to indicate whether the vehicle needs to be adjusted to avoid the obstacle.

[0053] For example, after the prediction of the future motion trajectory of the obstacle is completed, the second detection result is determined based on the future motion trajectory of the obstacle, and the second detection result is used to indicate whether the vehicle needs to be adjusted to avoid the obstacle. For example, the second detection result is determined based on the future motion trajectory of the obstacle, including: judging whether the lane where the vehicle is located intersects with the future motion trajectory of the obstacle, if the lane where the vehicle is located intersects with the future motion trajectory of the obstacle, the second detection result indicates that the vehicle needs to be adjusted to avoid the obstacle. If the lane where the vehicle is located does not intersect with the future motion trajectory of the obstacle, the second detection result indicates that the vehicle does not need to be adjusted to avoid the obstacle.

[0054] In step 206, in response to the second detection result indicating that the vehicle needs to be adjusted to avoid the obstacle, the driving state of the vehicle is adjusted based on the future motion trajectory of the obstacle.

[0055] In a possible implementation, when it is confirmed that the second detection result indicates that the vehicle needs to be adjusted to avoid the obstacle, the driving state of the vehicle is adjusted based on the future motion trajectory of the obstacle, including: in response to the future motion trajectory of the obstacle intersecting with the lane where the vehicle is located, and at least one lane on the left side or the right side of the vehicle not intersecting with the future motion trajectory of the obstacle and being passable, the vehicle is controlled to change lanes; in response to the future motion trajectory of the obstacle intersecting with the lane where the vehicle is located, and no lane on the left side and the right side of the vehicle not intersecting with the future motion trajectory of the obstacle, or the lane not intersecting with the future motion trajectory of the obstacle on the left side and the right side of the vehicle being impassable, a third detection result is obtained, the third detection result being used to indicate whether the vehicle reaches the position where the obstacle is located at this time when the vehicle travels at the current speed when the obstacle reaches the lane where the vehicle is located and is in the height range of the vehicle; and the driving state of the vehicle is adjusted based on the third detection result.

[0056] Exemplarily, when the future motion trajectory of the obstacle intersects with the lane where the vehicle is located, it is determined whether there is a lane on the left side or the right side of the vehicle, whether the lane on the left side or the right side intersects with the future motion trajectory of the obstacle, and whether the lane not intersecting with the future motion trajectory of the obstacle is passable. If there is at least one lane on the left side or the right side of the vehicle not intersecting with the future motion trajectory of the obstacle, and at least one lane not intersecting with the future motion trajectory of the obstacle is passable, it is indicated that the vehicle can change lanes, and the vehicle is controlled to change lanes, that is, the vehicle is adjusted to travel in the lane not intersecting with the future motion trajectory of the obstacle and being passable. If there is no lane on the left side and the right side of the vehicle not intersecting with the future motion trajectory of the obstacle, or the lane not intersecting with the future motion trajectory of the obstacle on the left side and the right side of the vehicle is impassable, a third detection result is obtained, the third detection result being used to indicate whether the vehicle reaches the position where the obstacle is located at this time when the vehicle travels at the current speed when the obstacle reaches the lane where the vehicle is located and is in the height range of the vehicle.

[0057] In a possible implementation, whether there is a lane on the left side or the right side of the vehicle and whether the lane on the left side or the right side of the vehicle is passable can be acquired by a navigation system. Optionally, the navigation system is installed on the vehicle, and the navigation system can collect traffic conditions around a road where the vehicle is located, including whether there is a lane on the left side or the right side of the vehicle, a driving direction of the lane on the left side or the right side of the vehicle, and a traffic control condition of the lane on the left side or the right side of the vehicle. Optionally, the lane changing of the vehicle can be controlled by controlling a steering system of the vehicle. The steering system of the vehicle includes a steering wheel, a steering gear, a steering knuckle, and a transmission rod, and controls a driving direction of the vehicle.

[0058] Optionally, the third detection result is acquired by: calculating, according to the predicted future motion trajectory of the obstacle, a preset time length required for the obstacle to enter a height range of the vehicle in the lane where the vehicle is located from a current position of the obstacle, and setting a position of the obstacle when the obstacle enters the height range of the vehicle in the lane where the vehicle is located as the first position. A second position reached by the vehicle at the preset time length after the vehicle continues to drive at a current speed is calculated, and the first position and the second position are compared. If a difference between the second position and the first position is less than a safety threshold, the third detection result indicates that, when the obstacle reaches the lane where the vehicle is located and is in the height range of the vehicle, the vehicle drives at the current speed to reach a position where the obstacle is located at this time. If the difference between the second position and the first position is greater than or equal to the safety threshold, and the second position is behind the first position, the third detection result indicates that, when the obstacle reaches the lane where the vehicle is located and is in the height range of the vehicle, the vehicle drives at the current speed and does not reach the position where the obstacle is located at this time. Exemplarily, the safety threshold can be set according to a safety distance required by the vehicle and the obstacle.

[0059] Exemplarily, after the third detection result is determined, the driving state of the vehicle is adjusted based on the third detection result. The adjustment includes: in response to the third detection result indicating that the vehicle reaches the position where the obstacle is located at this time when the obstacle reaches the height range of the vehicle, calculating an acceleration operation required for the vehicle to pass the position in advance before the obstacle reaches the position; when the acceleration operation is feasible, controlling the vehicle to drive according to the acceleration operation. When the acceleration operation is not feasible, calculating a deceleration operation required for the vehicle to pass the position after the obstacle reaches the position, and controlling the vehicle to drive according to the deceleration operation.

[0060] In a possible implementation, if the third detection result indicates that the obstacle reaches the height at which the vehicle is located, the vehicle reaches the position at which the obstacle is located at this time, the acceleration operation required for the vehicle to pass the position in advance before the obstacle reaches is calculated, including: calculating the distance from the current position of the vehicle to the first position, subtracting the safety passing time obtained by subtracting the margin time from the preset time to obtain the safety passing time, that is, it is considered that if the vehicle passes the first position by the margin time in advance of the preset time, the safety risk is small. Then divide the distance from the current position of the vehicle to the first position by the safety passing time to obtain the average speed required for the vehicle to pass the position in advance before the obstacle reaches. The maximum speed of the vehicle is calculated based on the average speed and the current speed (for example, assuming that the vehicle is uniformly accelerated or assuming that the vehicle is accelerated first and then uniformly accelerated, and then the acceleration performance and distance of the vehicle are introduced for calculation), and the maximum speed is compared with the speed limit requirement of the road. If the speed required for the advance passing position is less than or equal to the speed limit requirement of the road, it is indicated that the acceleration operation is feasible. If the maximum speed is greater than the speed limit requirement of the road, it is indicated that the acceleration operation is not feasible. The speed limit requirement of the road can be obtained through the navigation system.

[0061] Exemplarily, when the acceleration operation is feasible, the vehicle is controlled to travel according to the acceleration operation, that is, the vehicle is controlled to accelerate to the speed required for the advance passing position. When the acceleration operation is not feasible, the deceleration operation required for the vehicle to pass the position after the obstacle reaches is calculated, including: calculating the distance from the current position of the vehicle to the first position and subtracting the safety avoidance distance, and then dividing the distance from the current position of the vehicle to the first position and subtracting the safety avoidance distance by the preset time to obtain the average speed required for the vehicle to pass the position after the obstacle reaches. The vehicle is controlled to travel according to the deceleration operation. Alternatively, the power system of the vehicle can be controlled to control the acceleration or deceleration of the vehicle. The power system of the vehicle is composed of an engine, a transmission system and a brake system, and controls the travel speed of the vehicle. In a possible implementation, the safety avoidance distance can be set as the distance traveled by the vehicle at the current speed within the time interval from when the obstacle enters the lane in which the vehicle is located to when the obstacle leaves the lane in which the vehicle is located.

[0062] In a possible implementation, if the remaining vehicles appear within the first safety travel distance in front of the vehicle during the acceleration of the vehicle to pass the position in advance before the obstacle reaches, the driver is reminded through the central control large screen of the vehicle, and at the same time, if the emergency lane does not intersect with the future motion track of the obstacle, the vehicle can be automatically driven into the emergency lane and accelerated according to the original acceleration operation plan, so as to avoid the occurrence of danger due to the obstruction of the front vehicle and the failure to accelerate to pass the obstacle.

[0063] Exemplarily, if the remaining vehicle appears within the second safe driving distance behind the vehicle in the process that the vehicle decelerates through the position after the obstacle arrives, the driver is reminded by the central control large screen of the vehicle, the rear vehicle is prompted by the horn sound, and the vehicle continues to drive according to the original deceleration operation plan after driving into the emergency lane, so as to avoid the rear-end collision and other dangers caused by too close distance with the rear vehicle in the deceleration process. The first safe driving distance and the second safe driving distance can be set according to the current vehicle speed.

[0064] The embodiment of the present application determines whether there is an obstacle in front of the vehicle through the image detection result and the radar detection result, determines through the image detection result and the radar detection result, avoids the situation of missing detection due to the special shape of the special-shaped obstacle, and improves the accuracy of obstacle detection; in response to at least one of the radar detection result being that there is an obstacle or the image detection result being that there is an obstacle, it is determined whether the obstacle needs to be avoided based on the type and characteristics of the obstacle, so as to avoid the obstacle only when the characteristics of the obstacle meet the requirements of the type of obstacle that needs to be avoided, and improve the accuracy of vehicle obstacle avoidance. After confirming that the obstacle needs to be avoided, the future motion trajectory of the obstacle is predicted, the driving state of the vehicle is adjusted based on the future motion trajectory of the obstacle to avoid the obstacle, the flexibility of vehicle obstacle avoidance control is improved, and the reliability and safety of automatic driving are improved.

[0065] Referring to Figure 3 The embodiment of the present application provides an obstacle avoidance device of a vehicle, which comprises:

[0066] The acquisition module 301 is configured to acquire an image and a radar detection result in the driving direction of a road where the vehicle is located, and the radar detection result is used to indicate whether there is an obstacle within a first distance in front of the vehicle.

[0067] The detection module 302 is configured to detect the image in the driving direction of the road where the vehicle is located through an obstacle recognition model to obtain an image detection result, and the image detection result is used to indicate whether there is an obstacle within a second distance in front of the vehicle, the type of the obstacle, and the characteristics of the obstacle.

[0068] The first determination module 303 is configured to, in response to at least one of the radar detection result indicating that there is an obstacle within the first distance in front of the vehicle or the image detection result indicating that there is an obstacle within the second distance in front of the vehicle, determine a first detection result based on the type of the obstacle and the characteristics of the obstacle, and the first detection result is used to indicate whether the obstacle needs to be avoided.

[0069] The prediction module 304 is configured to, in response to the first detection result indicating that the obstacle needs to be avoided, predict a future motion trajectory of the obstacle.

[0070] The second determination module 305 is configured to determine a second detection result based on the future motion track of the obstacle, and the second detection result is used to indicate whether the vehicle needs to adjust the driving state to avoid the obstacle.

[0071] The adjustment module 306 is configured to, in response to the second detection result indicating that the vehicle needs to adjust the driving state to avoid the obstacle, adjust the driving state of the vehicle based on the future motion track of the obstacle.

[0072] In a possible implementation, the feature of the obstacle includes a size of the obstacle; and the first determination module 303 is configured to, in response to the type of the obstacle being a rockfall and the size of the obstacle being greater than a first threshold value, or the type of the obstacle being plastic and the size of the obstacle being greater than a second threshold value, determine that the first detection result indicates that the obstacle needs to be avoided.

[0073] In a possible implementation, the detection module 302 is further configured to establish an obstacle recognition model, an input of the obstacle recognition model being the image, and an output of the obstacle recognition model being the image detection result.

[0074] In a possible implementation, the prediction module 304 is configured to perform target tracking on the obstacle to obtain a motion state of the obstacle; analyze the motion state to determine an existing motion track and a current speed of the obstacle; and predict the future motion track of the obstacle based on the existing motion track and the current speed of the obstacle.

[0075] In a possible implementation, the adjustment module 306 is configured to, in response to the future motion track of the obstacle intersecting with a lane in which the vehicle is located, at least one lane on the left side or the right side of the vehicle not intersecting with the future motion track of the obstacle, and the at least one lane not intersecting with the future motion track of the obstacle being passable, control the vehicle to change lanes; in response to the future motion track of the obstacle intersecting with the lane in which the vehicle is located, no lane on the left side and the right side of the vehicle not intersecting with the future motion track of the obstacle, or the lane not intersecting with the future motion track of the obstacle on the left side and the right side of the vehicle being impassable, obtain a third detection result, the third detection result being used to indicate whether the vehicle reaches a position where the obstacle is located at this time when the vehicle travels at a current speed, when the obstacle reaches a height range of the vehicle; and adjust the driving state of the vehicle based on the third detection result.

[0076] In a possible implementation, the adjustment module 306 is configured to, in response to the third detection result indicating that the vehicle reaches the position where the obstacle is located at this time when the obstacle reaches the height range of the vehicle, calculate an acceleration operation required for the vehicle to pass through a position in advance before the obstacle reaches; and control the vehicle to travel according to the acceleration operation when the acceleration operation is feasible.

[0077] In a possible implementation, the adjusting module 306 is further configured to, when the acceleration operation is not feasible, calculate a deceleration operation required by the position after the obstacle arrives, and control the vehicle to travel according to the deceleration operation.

[0078] In a possible implementation, the device further includes a prompting module configured to, in response to the first detection result indicating that the obstacle needs to be avoided, prompt a driver of the vehicle that there is an obstacle in front of the vehicle.

[0079] The device determines whether there is an obstacle in front of the vehicle according to the image detection result and the radar detection result, determines whether there is an obstacle in front of the vehicle according to the image detection result and the radar detection result, avoids the case of missing detection due to the special shape of the special-shaped obstacle, and improves the accuracy of obstacle detection. In response to at least one of the radar detection result being that there is an obstacle or the image detection result being that there is an obstacle, whether the obstacle needs to be avoided is determined based on the type and characteristics of the obstacle, so that the obstacle is avoided only when the characteristics of the obstacle meet the requirements of the type of obstacle that needs to be avoided, and the accuracy of vehicle obstacle avoidance is improved. After confirming that the obstacle needs to be avoided, the future motion trajectory of the obstacle is predicted, the driving state of the vehicle is adjusted based on the future motion trajectory of the obstacle to avoid the obstacle, the flexibility of vehicle obstacle avoidance control is improved, and the reliability and safety of automatic driving are improved.

[0080] It should be noted that the device provided in the above embodiments is only exemplified by the above division of functional modules when realizing its functions. In actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the above described functions. In addition, the device and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process is described in the method embodiments, which will not be described here.

[0081] In an example embodiment, a non-transitory computer-readable storage medium is also provided, and the computer-readable storage medium stores at least one computer program. The at least one computer program is loaded and executed by a processor of a computer device, so that the computer implements any one of the above vehicle obstacle avoidance methods.

[0082] In a possible implementation, the above computer-readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a read-only compact disc (CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0083] In an example embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device performs any one of the above-mentioned vehicle obstacle avoidance methods.

[0084] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data for analysis, stored data, displayed data, etc.) and signals involved in the present application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions. For example, the image of the driving direction of the road where the vehicle is located and the radar detection result involved in the present application are obtained under full authorization.

[0085] It should be understood that "multiple" referred to herein refers to two or more. "And / or", which describes the association relationship of associated objects, means that there can be three relationships, for example, A and / or B can mean that there are three cases of A alone, A and B together, and B alone. The character " / " generally represents that the associated objects before and after it are in an "or" relationship.

[0086] It should be noted that the terms "first", "second", etc. (if any) in the specification and claims of the present application are used to distinguish similar objects, and do not necessarily describe a specific order or chronological order. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. The implementation described in the following example embodiments does not represent all implementations consistent with the present application. Rather, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0087] The above is only an example embodiment of the present application and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for obstacle avoidance by a vehicle, characterized in that, The method includes: The system collects images and radar detection results of the road in which the vehicle is traveling, and the radar detection results are used to indicate whether there are obstacles within a first distance in front of the vehicle. An obstacle recognition model is used to detect an image of the road in the direction of travel of the vehicle, and an image detection result is obtained. The image detection result is used to indicate whether there is an obstacle within a second distance in front of the vehicle, the type of the obstacle, and the characteristics of the obstacle. In response to the radar detection result indicating that there is an obstacle at a first distance in front of the vehicle or the image detection result indicating that there is an obstacle at a second distance in front of the vehicle, a first detection result is determined based on the type and characteristics of the obstacle, and the first detection result is used to indicate whether the obstacle needs to be avoided; In response to the first detection result indicating that the obstacle needs to be avoided, the future trajectory of the obstacle is predicted; A second detection result is determined based on the future trajectory of the obstacle, and the second detection result is used to indicate whether the vehicle needs to be adjusted to avoid the obstacle. In response to the second detection result indicating that the vehicle needs to avoid the obstacle, the vehicle's driving state is adjusted based on the future trajectory of the obstacle; Adjusting the vehicle's driving state based on the future trajectory of the obstacle includes: In response to the obstacle's future trajectory intersecting with the vehicle's lane, at least one lane on the left or right side of the vehicle does not intersect with the obstacle's future trajectory, and at least one lane that does not intersect with the obstacle's future trajectory is passable, the vehicle is controlled to change lanes. In response to the fact that the future trajectory of the obstacle intersects with the lane where the vehicle is located, and there are no lanes on the left and right sides of the vehicle that do not intersect with the future trajectory of the obstacle, or the lanes on the left and right sides of the vehicle that do not intersect with the future trajectory of the obstacle are impassable, a third detection result is obtained. The third detection result is used to indicate whether the vehicle has reached the current position of the obstacle when it is traveling at the current speed when the obstacle reaches the lane where the vehicle is located and is within the height range of the vehicle. The vehicle's driving status is adjusted based on the third detection result.

2. The method according to claim 1, characterized in that, The characteristics of the obstacle include the size of the obstacle; determining the first detection result based on the type of the obstacle and the characteristics of the obstacle includes: In response to the obstacle being a rockfall and its size being greater than a first threshold, or the obstacle being plastic and its size being greater than a second threshold, the first detection result is determined to indicate that the obstacle needs to be avoided.

3. The method according to claim 1, characterized in that, Before detecting the image of the vehicle's direction of travel on the road using the obstacle recognition model, the method further includes: establishing the obstacle recognition model, wherein the input of the obstacle recognition model is an image, and the output of the obstacle recognition model is the image detection result.

4. The method according to claim 1, characterized in that, The prediction of the future trajectory of the obstacle includes: The obstacle is tracked to obtain its motion state; The motion state is analyzed to determine the existing trajectory and current speed of the obstacle; Predict the future trajectory of the obstacle based on its existing trajectory and current speed.

5. The method according to claim 1, characterized in that, Adjusting the vehicle's driving state based on the third detection result includes: In response to the third detection result indicating that the obstacle has reached the height of the vehicle, the vehicle reaches the position of the obstacle at this time, and calculates the acceleration operation required to pass the position in advance before the obstacle arrives; When the acceleration operation is feasible, control the vehicle to move according to the acceleration operation.

6. The method according to claim 5, characterized in that, The calculation, after determining the acceleration required to pass the location before the obstacle arrives, also includes: When the acceleration operation is not feasible, calculate the deceleration operation required to pass through the position after the obstacle is reached, and control the vehicle to drive according to the deceleration operation.

7. The method according to claim 1, characterized in that, The method further includes: In response to the first detection result indicating that the obstacle needs to be avoided, the driver of the vehicle is alerted that there is an obstacle in front of the vehicle.

8. An obstacle avoidance device for a vehicle, characterized in that, The device includes: The acquisition module is used to acquire images and radar detection results of the road in which the vehicle is traveling. The radar detection results are used to indicate whether there are obstacles within a first distance in front of the vehicle. The detection module is used to detect images of the road in the direction of travel of the vehicle using an obstacle recognition model, and obtain image detection results. The image detection results are used to indicate whether there is an obstacle within a second distance in front of the vehicle, the type of the obstacle, and the characteristics of the obstacle. A first determining module is configured to determine a first detection result based on the type and characteristics of the obstacle in response to at least one of the radar detection result indicating that the obstacle exists at a first distance in front of the vehicle or the image detection result indicating that the obstacle exists at a second distance in front of the vehicle; the first detection result is used to indicate whether the obstacle needs to be avoided. The prediction module is used to predict the future trajectory of the obstacle in response to the first detection result indicating that the obstacle needs to be avoided; The second determining module is used to determine a second detection result based on the future trajectory of the obstacle, and the second detection result is used to indicate whether the vehicle needs to be adjusted to avoid the obstacle. An adjustment module is used to adjust the vehicle's driving state based on the future trajectory of the obstacle in response to the second detection result indicating that the vehicle needs to avoid the obstacle. The adjustment module is specifically used to control the vehicle to change lanes in response to the following: the future trajectory of the obstacle intersects with the lane where the vehicle is located; there is at least one lane on the left or right side of the vehicle that does not intersect with the future trajectory of the obstacle; and at least one lane that does not intersect with the future trajectory of the obstacle is passable. In response to the fact that the future trajectory of the obstacle intersects with the lane where the vehicle is located, and there are no lanes on the left and right sides of the vehicle that do not intersect with the future trajectory of the obstacle, or the lanes on the left and right sides of the vehicle that do not intersect with the future trajectory of the obstacle are impassable, a third detection result is obtained. The third detection result is used to indicate whether the vehicle has reached the current position of the obstacle when it is traveling at the current speed when the obstacle reaches the lane where the vehicle is located and is within the height range of the vehicle. The vehicle's driving status is adjusted based on the third detection result.

9. A non-transitory computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one computer program, which is loaded and executed by a processor to enable the computer to implement the obstacle detection method as described in any one of claims 1 to 7.

Images