Vehicle avoidance method and device, electronic equipment and storage medium

Abstract

The application provides a vehicle avoidance method and device, an electronic device and a storage medium, wherein the vehicle avoidance method comprises: obtaining state information and surrounding environment information of a vehicle; determining an avoidance target based on the state information and the surrounding environment information of the vehicle; when a driving scene of the vehicle is a target scene, controlling the vehicle to exit an avoidance mode, otherwise, controlling the vehicle to avoid the avoidance target based on the state information and the surrounding environment information of the vehicle, wherein the target scene comprises a sandwiching scene on both sides of the vehicle. The application at least improves driving safety of the vehicle.

Description

Technical Field

[0001] This application relates to the field of vehicle control, and more specifically, to a vehicle avoidance method, device, electronic device, and storage medium. Background Technology

[0002] Autonomous obstacle avoidance is widely used in autonomous vehicles. However, current intelligent obstacle avoidance decision-making schemes that enable autonomous obstacle avoidance reduce vehicle driving safety. Summary of the Invention

[0003] The purpose of this application is to provide a vehicle avoidance method, device, electronic device, and storage medium to at least improve vehicle driving safety.

[0004] In a first aspect, the present invention provides a vehicle avoidance method, the method comprising:

[0005] Obtain vehicle status information and surrounding environment information;

[0006] The avoidance target is determined based on the vehicle's status information and the surrounding environment information;

[0007] When the vehicle's driving scenario is the target scenario, the vehicle is controlled to exit the avoidance mode; otherwise, based on the vehicle's status information and the surrounding environment information, the vehicle is controlled to avoid the avoidance target. The target scenario includes a scenario where the vehicle is flanked by both sides.

[0008] The method of the first aspect of this application can avoid a target when it is identified. At the same time, by controlling the vehicle to exit the avoidance mode in the scenario of being surrounded by both sides, it can prevent the vehicle from swaying left and right due to frequent avoidance of avoidance targets located on the left and right sides of the vehicle in the scenario of being surrounded by both sides, thereby avoiding the reduction of vehicle driving safety due to swaying left and right, and thus improving vehicle driving safety.

[0009] In an optional implementation, the vehicle's surrounding environment information includes the vehicle speed and position of surrounding moving targets, and the vehicle's state information includes the vehicle speed and position.

[0010] And, the step of determining the avoidance target based on the vehicle's state information and the surrounding environment information includes:

[0011] The longitudinal distance of the vehicle relative to the surrounding moving targets is obtained based on the position of the vehicle and the position of the surrounding moving targets;

[0012] The speed of the vehicle relative to the surrounding moving targets is obtained based on the speed of the vehicle and the speed of the surrounding moving targets;

[0013] Based on the longitudinal distance of the vehicle relative to the surrounding moving target and the speed of the vehicle relative to the surrounding moving target, the longitudinal collision time distance between the vehicle and the surrounding moving target is obtained;

[0014] When the longitudinal collision distance is positive, the surrounding moving target is identified as the avoidance target.

[0015] This optional implementation can improve the rationality and effectiveness of the vehicle's avoidance maneuvers, thereby improving vehicle comfort. The longitudinal collision distance between the vehicle and the surrounding moving target is determined based on the relative distance and relative speed between the vehicles. This means that the determination of the avoidance target is related not only to the relative distance between the vehicles but also to the vehicle's speed relative to the surrounding moving target, reducing the probability of misjudging the avoidance target. This, in turn, reduces the probability of misjudging the target and triggering the vehicle's avoidance function, thus improving the rationality and effectiveness of the vehicle's avoidance maneuvers and ultimately enhancing vehicle comfort.

[0016] In an optional implementation, the surrounding environment information further includes lane information;

[0017] Furthermore, before determining the avoidance target based on the vehicle's state information and the surrounding environment information, the method further includes:

[0018] The width of the lane in which the vehicle is located and the confidence level of the lanes surrounding the vehicle are determined based on the lane information.

[0019] When the width of the lane where the vehicle is located is not zero, and the confidence level of the lanes surrounding the vehicle is greater than or equal to a preset confidence threshold, determine whether the vehicle speed is greater than or equal to a preset speed threshold.

[0020] When the vehicle's speed is greater than or equal to a preset speed threshold, the process of determining the avoidance target based on the vehicle's state information and the surrounding environment information is triggered.

[0021] This optional implementation can further improve driving safety and comfort by verifying lane confidence, lane width, and vehicle speed.

[0022] In an optional implementation, after determining the avoidance target based on the vehicle's state information and the surrounding environment information, the method further includes:

[0023] Based on the position of the avoidance target and the position of the vehicle, determine whether the avoidance target appears on both sides of the vehicle;

[0024] When the avoidance target appears on both sides of the vehicle, it is determined whether the longitudinal distance between the vehicle and each avoidance target is less than or equal to a preset longitudinal distance threshold.

[0025] When the longitudinal distance between the vehicle and each of the avoidance targets is less than or equal to the preset longitudinal distance threshold, the driving scenario of the vehicle is a scenario where the vehicle is flanked by both sides.

[0026] This optional implementation can detect and identify a vehicle being flanked from both sides by detecting the location of the target to be avoided and the longitudinal distance between the target and the vehicle.

[0027] In an optional implementation, the target scenario further includes a small-radius curve scenario;

[0028] Furthermore, the method further includes:

[0029] When the avoidance target appears on both sides of the vehicle, it is determined whether the lane where the vehicle is located is a curve based on the lane information, and whether the radius of the curve is less than or equal to a preset radius threshold.

[0030] If the lane where the vehicle is located is the curve, and the radius of the curve is less than or equal to the preset radius threshold, then the vehicle's driving scenario is the small-radius curve scenario.

[0031] This application can identify small-radius curve scenarios and prevent the avoidance function from being triggered in such scenarios, thereby further improving driving safety.

[0032] In an optional implementation, the target scenario further includes scenarios where the longitudinal collision distance is relatively small when the target appears;

[0033] Furthermore, the method further includes:

[0034] Based on the speed of the target being avoided and the position of the target being avoided, it is determined whether the target being avoided slows down after passing behind the vehicle to the front of the vehicle, or whether the target being avoided runs parallel to the vehicle when cutting in.

[0035] When the vehicle decelerates after the avoidance target passes behind the vehicle and moves in front of the vehicle, or when the avoidance target cuts in parallel with the vehicle, the driving scenario of the vehicle is a scenario where the longitudinal collision distance is small when the target appears.

[0036] This optional implementation can prevent the avoidance function from being triggered in scenarios with a small longitudinal collision distance, thereby further improving driving safety.

[0037] In an optional implementation, the target scenario further includes a vehicle lane-changing scenario;

[0038] Furthermore, the method further includes:

[0039] Based on the vehicle's position and the lane information, determine whether the vehicle is in the process of changing lanes;

[0040] When the vehicle is in the process of changing lanes, the driving scenario of the vehicle is a lane-changing scenario.

[0041] This optional implementation can prevent the avoidance function from being triggered in the lane-changing scenario by recognizing the lane-changing scenario of the vehicle, thereby further improving driving safety.

[0042] In an optional implementation, controlling the vehicle to avoid the avoidance target based on the vehicle's state information and the surrounding environment information includes:

[0043] The lateral distance of the avoidance target relative to the vehicle is determined based on the position of the vehicle and the position of the avoidance target;

[0044] The lateral velocity of the target relative to the ground is determined based on the velocity of the target being avoided;

[0045] Obtain the distance limit values ​​between the vehicle and the left and right lane lines;

[0046] When the longitudinal collision distance between the vehicle and the surrounding moving target is within a preset time distance value range, and the absolute value of the lateral distance between the avoidance target and the vehicle is within a preset distance value range, and the lateral velocity of the avoidance target relative to the ground is within a preset speed value range, and the avoidance target does not change within the system cycle, an avoidance activation command is determined and an avoidance offset is determined based on the width of the lane and the distance limit values ​​between the vehicle and the left and right lane lines;

[0047] The vehicle is controlled to avoid the target based on the avoidance offset.

[0048] This optional implementation can further use whether the avoidance target changes as the activation condition during the avoidance function activation process, thereby avoiding repeated advance and retreat avoidance functions and further improving comfort.

[0049] In an optional implementation, the target to be avoided includes a two-wheeled vehicle.

[0050] This optional implementation can target two-wheeled vehicles, thus increasing the range of the obstacle avoidance function.

[0051] In a second aspect, the present invention provides a vehicle avoidance device, the device comprising:

[0052] The acquisition module is used to acquire vehicle status information and surrounding environment information;

[0053] The determination module is used to determine the avoidance target based on the vehicle's state information and the surrounding environment information;

[0054] The control module is used to control the vehicle to exit the avoidance mode when the driving scenario of the vehicle is the target scenario; otherwise, it controls the vehicle to avoid the avoidance target based on the vehicle's status information and the surrounding environment information, wherein the target scenario includes a scenario where the vehicle is surrounded by both sides.

[0055] The device of the second aspect of this application can avoid a target when it is detected. At the same time, by controlling the vehicle to exit the avoidance mode in a scenario where it is surrounded by two sides, it can prevent the vehicle from swaying left and right due to frequent avoidance of targets located on the left and right sides of the vehicle in such a scenario, thereby avoiding the reduction of vehicle driving safety by swaying left and right.

[0056] Thirdly, the present invention provides an electronic device, comprising:

[0057] Processor; and

[0058] The memory is configured to store machine-readable instructions that, when executed by the processor, perform the vehicle avoidance method as described in any of the foregoing embodiments.

[0059] The electronic device of the third aspect of this application, by executing a vehicle avoidance method, can avoid a target when it is identified. Simultaneously, by controlling the vehicle to exit the avoidance mode in a scenario where it is flanked from both sides, it can prevent the vehicle from frequently avoiding targets located on the left and right sides of the vehicle in such a scenario, thus avoiding lateral swaying and reducing vehicle driving safety, thereby improving vehicle driving safety.

[0060] Fourthly, the present invention provides a storage medium storing a computer program, the computer program being executed by a processor as described in any of the foregoing embodiments of the vehicle avoidance method.

[0061] The storage medium of the fourth aspect of this application, by executing a vehicle avoidance method, can avoid a target when it is identified. Simultaneously, by controlling the vehicle to exit the avoidance mode in a scenario where it is flanked from both sides, it can prevent the vehicle from frequently avoiding targets located on the left and right sides of the vehicle in such scenarios, thus preventing lateral swaying and reducing vehicle driving safety. Attached Figure Description

[0062] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0063] Figure 1 This is a schematic flowchart of a vehicle avoidance method provided in an embodiment of this application;

[0064] Figure 2 This is a schematic diagram of the structure of a vehicle avoidance device provided in an embodiment of this application;

[0065] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application;

[0066] Figure 4 This is a schematic diagram of a sensing target provided in an embodiment of this application. Detailed Implementation

[0067] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.

[0068] Currently, existing intelligent obstacle avoidance decision-making schemes that enable autonomous obstacle avoidance actually reduce vehicle safety. This is because these schemes fail to consider the specific scenario conditions under which the vehicle is triggered for autonomous obstacle avoidance. In such specific scenarios, triggering autonomous obstacle avoidance not only fails to reduce safety but actually exacerbates it. Specifically, when there are obstacles on both sides of the vehicle, existing intelligent obstacle avoidance decision-making schemes first avoid the obstacle on the left, causing the vehicle to veer to the right. After successfully avoiding the obstacle on the right, the vehicle will then veer to the left, or vice versa. This results in the vehicle frequently switching between left and right, making its trajectory unstable and increasing the risk of accidents.

[0069] Based on this, this application provides a vehicle avoidance method, apparatus, device, and storage medium, which can avoid the avoidance target when it is identified. Simultaneously, by controlling the vehicle to exit the avoidance mode in a scenario where it is surrounded from both sides, it can prevent the vehicle from frequently avoiding avoidance targets located on the left and right sides of the vehicle in such scenarios, thus avoiding lateral swaying and reducing vehicle driving safety, thereby improving vehicle driving safety.

[0070] Please see Figure 1 , Figure 1 This is a flowchart illustrating a vehicle avoidance method provided in an embodiment of this application. Figure 1 As shown, the vehicle avoidance method of this application embodiment includes the following steps:

[0071] 101. Obtain vehicle status information and surrounding environment information;

[0072] 102. Determine the avoidance target based on the vehicle's status information and surrounding environment information;

[0073] 103. When the vehicle's driving scenario is the target scenario, control the vehicle to exit the avoidance mode; otherwise, control the vehicle to avoid the target based on the vehicle's status information and surrounding environment information. The target scenario includes scenarios where the vehicle is flanked by both sides.

[0074] In the above steps, steps 101 and 102 determine the avoidance target based on the vehicle's state information and surrounding environment information, giving the vehicle's avoidance function a clear avoidance target, which can serve as one of the triggering conditions for the vehicle's avoidance function. Further, step 103 allows the vehicle to avoid the avoidance target. Furthermore, by controlling the vehicle to exit avoidance mode or not trigger avoidance mode in scenarios where it is flanked from both sides, the vehicle can be prevented from frequently avoiding avoidance targets located on its left and right sides in such scenarios, thus avoiding lateral swaying and improving vehicle driving safety.

[0075] In this embodiment, the vehicle status information in step 101 refers to data characterizing the current state of the vehicle. This vehicle status information includes, but is not limited to, the vehicle's speed, position, model, acceleration, weight, torque, and wheel torque difference. Furthermore, the acquisition of vehicle status information can be determined by the vehicle control terminal. The vehicle is equipped with sensors that detect various status data. These sensors transmit real-time detection results to the vehicle control terminal, which can then determine the status information of one or more vehicles based on the detection results of one sensor or by fusing the detection results of multiple sensors.

[0076] Furthermore, vehicle status information can also be determined by fusing detection results from external devices. For example, the vehicle control terminal can fuse detection results from one or more of its own sensors and positioning data sent by a satellite communication terminal to comprehensively determine the vehicle's status information. The vehicle control terminal can also fuse detection results from one or more of its own sensors with data sent by surrounding vehicles to obtain the vehicle's status information.

[0077] Furthermore, the sensors used to detect vehicle status information include, but are not limited to, temperature sensors, infrared sensors, inertial detection units, and lidar.

[0078] In this embodiment of the application, for step 101, the surrounding environment information refers to the detection information of objects around the vehicle, wherein the surrounding objects include lanes, moving targets, objects in fixed positions, etc. Sensors used to detect the surrounding environment information include, but are not limited to, temperature sensors, infrared sensors, inertial detection units, and lidar.

[0079] In this embodiment, for step 102, the avoidance target refers to an object that may collide with the vehicle or cause the vehicle to collide with other targets. Further, the avoidance target can include objects that are moving or objects that are located in a fixed position. In this embodiment, considering the significant threat that moving objects pose to vehicle safety, moving objects are preferably chosen as avoidance targets. However, this is only a preferred option and does not preclude other types of objects from being considered avoidance targets.

[0080] In this application embodiment, for step 102, one implementation of determining the avoidance target based on the vehicle's state information and surrounding environment information can be as follows: first, determine the vehicle's position based on the vehicle's state information; then, determine whether there is a moving object around the vehicle based on the surrounding environment information; if there is a moving object, determine the position of the moving object based on the surrounding environment information; finally, determine the longitudinal distance between the vehicle and the moving object based on the position of the moving object and the vehicle's position; if the longitudinal distance is less than a threshold, it is considered that the moving object is too close to the vehicle and needs to be avoided, that is, the moving object is determined as the avoidance target.

[0081] In this embodiment of the application, regarding step 102, in addition to determining whether the moving object is a target to be avoided based on the longitudinal distance between the vehicle and the moving object, another implementation method for determining the target to be avoided based on the relative speed between the vehicle and the moving object, i.e., based on the vehicle's state information and the surrounding environment information, is as follows: first, determine the vehicle's speed based on the vehicle's state information; then, determine whether there is a moving object around the vehicle based on the surrounding environment information; if there is a moving object, determine the speed of the moving object based on the surrounding environment information; finally, determine whether the speed of the moving object is greater than the speed of the vehicle based on the speed of the moving object and the speed of the vehicle; if so, it is considered that the moving object is too close to the vehicle and needs to be avoided, i.e., the moving object is determined as a target to be avoided.

[0082] In this embodiment, step 103, "avoidance mode," refers to the working state in which the vehicle implements autonomous avoidance measures, also known as the vehicle's avoidance function. Furthermore, in avoidance mode, when the vehicle detects a target to avoid, it can autonomously operate the steering wheel and brake pedal without the driver operating the steering wheel or brake pedal. Furthermore, avoidance modes include, but are not limited to, different types of avoidance modes such as avoidance modes for highways and avoidance modes for urban roads.

[0083] In this embodiment of the application, the driving scenario of the vehicle in step 103 refers to the vehicle driving situation reflected by the current state of the vehicle and the current state of the surrounding environment. The driving scenario of the vehicle can be determined based on the current state of the vehicle, or the vehicle driving situation reflected by the current state of the surrounding environment, or by combining the vehicle driving situation reflected by the current state of the surrounding environment and the current state of the vehicle.

[0084] In this embodiment of the application, step 103, the target scenario, refers to the driving scenario in which the vehicle needs to exit the avoidance mode or not trigger the avoidance mode, also known as a special scenario. For example, when the vehicle's driving scenario is that there are avoidance targets on both the left and right sides of the vehicle, in order to ensure that the vehicle's avoidance function reduces the vehicle's driving safety in this driving scenario, the vehicle needs to exit the avoidance mode in this driving scenario, that is, the vehicle's driving scenario is the target scenario.

[0085] Furthermore, the target scenarios include, but are not limited to, scenarios where the vehicle needs to exit the avoidance mode or not trigger the avoidance mode, such as scenarios where the vehicle is flanked by both sides, scenarios where the radius of the curve is small, scenarios where the longitudinal collision distance is small when the target appears, and scenarios where the vehicle changes lanes. The determination of each driving scenario can be based on the vehicle's status information and the surrounding environment information. For example, the driving scenario of the vehicle can be determined based on the lane information in the surrounding environment information and the vehicle speed information in the vehicle's status information.

[0086] In some specific scenarios, relying solely on the longitudinal distance between the vehicle and surrounding moving objects can reduce the user's riding comfort. For example, when the longitudinal distance between the vehicle and surrounding moving objects is less than a judgment threshold, but the vehicle's speed is slower than the speed of the surrounding moving objects, the vehicle is not at risk of colliding with the surrounding moving objects. On the contrary, because the speed of the surrounding moving objects is higher than the speed of the vehicle, the longitudinal distance between the vehicle and surrounding moving objects gradually increases. If the surrounding moving object is used as the avoidance target and the avoidance function is triggered, it will cause the vehicle to brake suddenly or turn the steering wheel sharply when there is no risk of collision, thereby reducing the comfort of the vehicle occupants, or making the user feel that the distance between them and the vehicle in front is too large.

[0087] For this specific scenario, in one optional implementation, the step of determining the avoidance target based on the vehicle's state information and surrounding environment information is as follows:

[0088] The longitudinal distance of the vehicle relative to the surrounding moving targets is obtained based on the vehicle's position and the positions of the surrounding moving targets. The surrounding environment information of the vehicle includes the speed and position of the surrounding moving targets, and the vehicle's state information includes the vehicle's speed and position.

[0089] The vehicle's speed relative to the surrounding moving targets is obtained based on the vehicle's speed and the speeds of surrounding moving targets.

[0090] Based on the longitudinal distance of the vehicle relative to the surrounding moving targets and the speed of the vehicle relative to the surrounding moving targets, the longitudinal collision time between the vehicle and the surrounding moving targets is obtained;

[0091] When the distance to the longitudinal collision is positive, surrounding moving targets are identified as avoidable targets.

[0092] This optional implementation can improve the rationality and effectiveness of vehicle avoidance maneuvers, thereby improving vehicle comfort. The longitudinal collision distance between the vehicle and surrounding moving targets is determined based on the relative distance and relative speed between the vehicles. This means that target determination considers not only the relative distance but also the vehicle's speed relative to the surrounding moving target, reducing the probability of misjudging the target and thus lowering the probability of false triggering of the vehicle's avoidance function. In other words, it improves the rationality and effectiveness of the vehicle's avoidance maneuvers, thereby enhancing vehicle comfort. For example, when the longitudinal distance between the vehicle and a surrounding moving object is less than a judgment threshold, but the vehicle's speed is slower than the surrounding moving object's speed, the surrounding moving object can be excluded from the avoidance target. This prevents the vehicle from braking suddenly or turning sharply when there is no collision risk, further improving the rationality and effectiveness of the vehicle's avoidance maneuvers and ultimately enhancing the comfort of vehicle occupants.

[0093] In the above implementation, the positions of the vehicle and the surrounding moving targets can be determined using satellite positioning data.

[0094] In the above embodiments, the longitudinal distance of the vehicle relative to the surrounding moving targets refers to the distance difference in the longitudinal direction calculated based on the position of the vehicle and the position of the surrounding moving objects.

[0095] In the above embodiments, the speed of the vehicle relative to the surrounding moving target refers to the difference between the vehicle's speed and the speed of the surrounding moving target, also known as the vehicle's relative speed. Further, one implementation method for obtaining the longitudinal collision time distance between the vehicle and the surrounding moving target based on the vehicle's longitudinal distance relative to the surrounding moving target and the vehicle's speed relative to the surrounding moving target could be:

[0096] A = B / C;

[0097] Where A represents the longitudinal collision time between the vehicle and the surrounding moving target, B represents the relative longitudinal distance between the vehicle and the surrounding moving target, and C represents the speed of the vehicle relative to the surrounding moving target.

[0098] Furthermore, the calculated longitudinal collision time distance between the vehicle and surrounding moving targets is compared with zero. If the longitudinal collision time distance between the vehicle and surrounding moving targets is greater than zero (i.e., a positive value), it indicates a collision risk at the current speed and longitudinal distance, and the surrounding moving objects are considered avoidance targets. If the longitudinal collision time distance between the vehicle and surrounding moving targets is less than zero (i.e., a negative value), it indicates no collision risk at the current speed and longitudinal distance, and the surrounding moving objects do not need to be identified as avoidance targets.

[0099] It should be noted that the sign of the longitudinal collision distance between surrounding moving targets is determined by the vehicle's speed relative to the surrounding moving targets. For example, if the vehicle's speed relative to the surrounding moving targets is negative (i.e., the vehicle's speed is slower than the surrounding moving targets), the longitudinal collision distance between the surrounding moving targets is negative, and in this case, the surrounding moving targets do not need to be considered as avoidable targets. However, if the vehicle's speed relative to the surrounding moving targets is positive (i.e., the vehicle's speed is faster than the surrounding moving targets), the longitudinal collision distance between the surrounding moving targets is positive, and in this case, the vehicle may catch up with and collide with the surrounding moving targets.

[0100] It should be noted that the surrounding moving targets are preferably moving targets that are in the same lane as the vehicle.

[0101] In some specific scenarios, vehicles travel on roads without lane markings, such as mixed lanes—roads that can be used by both pedestrians and motor vehicles. On roads without lane markings, if the obstacle avoidance function is triggered, the lack of lane markings prevents the function from determining the appropriate avoidance space, potentially leading to safety issues. For example, due to the absence of lane markings, the vehicle's avoidance function may not be able to determine whether it has properly avoided an obstacle, or it may mistakenly identify moving objects around the vehicle as moving objects in its own lane instead of those of adjacent vehicles, thus executing incorrect avoidance measures and causing further safety problems.

[0102] In this scenario, in an optional implementation, before determining the avoidance target based on the vehicle's state information and surrounding environment information, the method of this application embodiment further includes the following steps:

[0103] The width of the lane in which the vehicle is located and the confidence level of the lanes surrounding the vehicle are determined based on lane information, where the surrounding environment information also includes lane information.

[0104] If the width of the lane in which the vehicle is located is not zero, and the confidence level of the lanes surrounding the vehicle is greater than or equal to the preset confidence threshold, determine whether the vehicle speed is greater than or equal to the preset speed threshold.

[0105] When the vehicle speed is greater than or equal to a preset speed threshold, the system will trigger the determination of the avoidance target based on the vehicle's status information and the surrounding environment information.

[0106] Therefore, in the above implementation, the next action is triggered only when the width of the lane where the vehicle is located is not zero. When the width of the lane where the vehicle is located is zero, it means that it cannot be determined that the lane where the vehicle is located is a dedicated motor vehicle lane. Therefore, the action of determining the avoidance target based on the vehicle's status information and the surrounding environment information is not triggered. This can prevent the avoidance function from making incorrect judgments based on lane markings and executing incorrect avoidance measures, thereby avoiding traffic accidents caused by incorrect avoidance measures and improving the driving safety of the vehicle.

[0107] Furthermore, lanes vary in width, and some lanes have gradually been repurposed for other uses, such as parking. If these special lanes are used as yield lanes for vehicles, it can lead to traffic accidents. Similarly, narrow lanes like non-motorized vehicle lanes, if used as yield lanes for vehicles, may cause vehicles to enter these lanes during yielding, increasing the risk of collisions with pedestrians and resulting in accidents. Therefore, it is necessary to further assess the confidence level of the surrounding lanes. The next step should only be performed if the confidence level of the surrounding lanes is greater than or equal to a preset confidence threshold. If the confidence level does not meet this threshold, the yield target determination based on vehicle status and surrounding environment information should not be performed. This prevents unreliable surrounding lanes from being used as yield spaces for the yielding function, thus avoiding traffic accidents caused by incorrect determination of the surrounding lanes.

[0108] Furthermore, in some scenarios, vehicles travel at low speeds. This provides drivers with sufficient reaction time to mitigate the risk of collisions with surrounding objects. Even if a collision does occur, the slow speed results in a relatively minor accident. For example, a vehicle traveling at 5 km / h through a congested market street is less likely to collide with a pedestrian than an obstacle avoidance system that accelerates and changes lanes to avoid collisions with surrounding objects. Therefore, the aforementioned optional implementation, based on the vehicle's lane width and the confidence level of surrounding lanes, further determines whether the vehicle's speed is greater than or equal to a preset speed threshold. Only when the vehicle's speed is greater than or equal to the preset speed threshold is the process of determining an avoidance target based on the vehicle's state information and the surrounding environment information initiated. If the vehicle's speed does not meet the preset speed threshold, the process of determining an avoidance target based on the vehicle's state information and the surrounding environment information is not triggered, thereby further improving driving safety.

[0109] Furthermore, when the vehicle is moving slowly, by avoiding triggering the obstacle avoidance function, the user's ride comfort can be improved by avoiding measures such as emergency acceleration and lane changing implemented by the obstacle avoidance function.

[0110] In the above implementation, the confidence level of the surrounding lanes of a vehicle indicates the reliability of whether the surrounding lanes of the vehicle can be used as a passing lane for the vehicle. If the confidence level of the surrounding lanes of a vehicle is greater than or equal to a preset confidence threshold, the surrounding lanes are considered to be highly reliable and can be used as passing lanes for the vehicle. If the confidence level of the surrounding lanes of a vehicle is not greater than or equal to the preset confidence threshold, the surrounding lanes are considered to be unreliable and cannot be used as passing lanes for the vehicle.

[0111] Furthermore, the preset confidence threshold can be 70% or 80%, and this embodiment of the application does not limit this. In addition, the confidence level of a lane can be determined by comprehensively considering factors such as the width of surrounding lanes and signage. For example, when the width of surrounding lanes is less than 2.8m, the confidence level of that lane can be determined to be 0. Furthermore, when multiple factors are involved, each factor corresponds to a weight value. By weighting these weight values, a preset confidence threshold can be obtained by comprehensively considering multiple factors.

[0112] Furthermore, the preset speed threshold refers to the minimum speed required to trigger the avoidance function. The preset speed threshold can be 30 kilometers per hour or 25 kilometers per hour, and this embodiment of the application does not limit it.

[0113] Therefore, this optional implementation method can further improve driving safety and comfort by verifying lane confidence, lane width, and vehicle speed.

[0114] In some specific scenarios, there are targets to avoid on both sides of the vehicle. If the avoidance function is triggered at this time, the vehicle's avoidance function will first avoid the target on one side of the vehicle. Then, because the target on the other side also triggers the vehicle's avoidance function, the vehicle will frequently avoid left and right while the targets on both sides of the vehicle are still present. This causes the vehicle to sway left and right, which reduces the vehicle's driving safety.

[0115] For this scenario, it is necessary to identify driving scenarios where there are targets to avoid on both the left and right sides of the vehicle, i.e., identifying scenarios where the vehicle is flanked by both sides. Therefore, in an optional implementation, after determining the avoidance targets based on the vehicle's state information and surrounding environment information, the method of this application embodiment further includes the following steps:

[0116] Determine whether there are targets to avoid on both sides of the vehicle based on the location of the target to avoid and the vehicle's location;

[0117] When there are targets to avoid on both sides of the vehicle, determine whether the longitudinal distance between the vehicle and each target to avoid is less than or equal to a preset longitudinal distance threshold.

[0118] When the longitudinal distance between the vehicle and each target to be avoided is less than or equal to a preset longitudinal distance threshold, the vehicle's driving scenario is a scenario where the vehicle is flanked by both sides.

[0119] This optional implementation can detect and identify the scenario of vehicles being sandwiched on both sides by detecting the position of the target to be avoided and the longitudinal distance between the target and the vehicle. Then, the scenario of vehicles being sandwiched on both sides can be used as one of the target scenarios, so that the avoidance function is not triggered when the vehicle is sandwiched on both sides, thereby avoiding the vehicle from swaying frequently from side to side and improving vehicle driving safety.

[0120] In the above implementation, the position of the avoidance target is compared with the position of the vehicle. If one avoidance target is located to the left of the vehicle's position and the other avoidance target is located to the right of the vehicle's position, it is determined that there are avoidance targets on both sides of the vehicle.

[0121] In the above embodiments, the preset longitudinal distance threshold can be 2 meters or 3 meters. This application embodiment does not limit this. The preset longitudinal distance threshold can refer to the distance that would cause the vehicle to perform another avoidance measure when the time between the last avoidance operation and the last avoidance operation is less than a specified time interval.

[0122] In some specific scenarios, there are targets to avoid on both sides of the vehicle, but the condition that the longitudinal distance between the vehicle and each target is less than or equal to a preset longitudinal distance threshold is not met, thus preventing the vehicle from swaying frequently from side to side. However, the vehicle may be turning in a very small curve. In this case, triggering the avoidance function could easily cause the vehicle to fly off the curve, resulting in a traffic accident.

[0123] In this scenario, the method of this application embodiment also needs to include small-radius curve scenarios as one of the target scenarios. Specifically, the specific steps for identifying small-radius curve scenarios include:

[0124] When there are targets to avoid on both sides of the vehicle, the system determines whether the lane in which the vehicle is located is a curve based on lane information, and whether the radius of the curve is less than or equal to a preset radius threshold.

[0125] If the lane in which the vehicle is located is a curve, and the radius of the curve is less than or equal to a preset radius threshold, the vehicle's driving scenario is a small-radius curve scenario.

[0126] Therefore, this application can further determine whether the lane is curved and whether the radius of the curve is less than or equal to a preset radius threshold when there are avoidance targets on both sides of the vehicle. If this condition is met, the vehicle's driving scenario is a small radius curve scenario, and the small radius curve scenario is taken as one of the target scenarios. Thus, the avoidance function is not triggered in the small radius curve scenario, which can prevent the avoidance function from triggering in the small radius curve scenario and causing driving safety problems, thereby further improving driving safety.

[0127] In the above implementation, lane information includes the vehicle's curvature, wherein the curvature of the vehicle can be used to determine whether the lane is a curve. Furthermore, the curvature of the vehicle can be used to determine the radius of the curve.

[0128] In the above implementation, the preset radius threshold refers to the radius threshold at which a vehicle might run off the curve when attempting to avoid it. For example, if a vehicle performs the avoidance function when the curve radius is less than or equal to R, it will run off the curve, and the curve radius R is the preset radius threshold. It should be noted that the smaller the curve radius, the sharper the curve generally is, and at the same vehicle speed, it is easier for a vehicle to run off the curve.

[0129] In certain specific scenarios, if a moving object has just cut in front of a vehicle and is running parallel to it, or has just cut in front of the vehicle and is slowing down, triggering the obstacle avoidance function could lead to a traffic accident. Typically, when a moving object cuts in front of a vehicle and is running parallel to it, or has just cut in front of the vehicle and is slowing down, it means that the moving object is traveling at a speed at least equal to that of the vehicle. In this case, if the longitudinal distance between the moving object and the vehicle is small, if the obstacle avoidance function is triggered, the function will either control the vehicle to accelerate to change lanes or overtake, or control the vehicle to brake suddenly to decelerate. For the measure of accelerating to change lanes or overtaking, the vehicle needs to accelerate significantly. As a result, when the longitudinal distance between the vehicle and the moving object is small, it is more likely to collide with the moving object. If the vehicle brakes suddenly to decelerate, the speed reduction is greater due to the small longitudinal distance between the vehicle and the moving object, making it more likely to collide with vehicles coming from behind. Therefore, when a moving object suddenly appears alongside a vehicle, or when a moving object suddenly appears and passes in front of the vehicle and slows down, and the longitudinal distance between the moving object and the vehicle is small, the vehicle's speed will change drastically compared to normal avoidance, making it more likely to cause a traffic accident.

[0130] In response to the above scenarios, in an optional implementation, scenarios where the longitudinal collision distance is relatively small when the target appears are considered as one of the target scenarios to avoid triggering the avoidance function in such scenarios. Accordingly, the method of this application embodiment further includes the following steps:

[0131] Based on the speed and position of the target being avoided, determine whether the target being avoided slows down after passing from behind the vehicle to in front of the vehicle, or whether the target being avoided runs parallel to the vehicle when cutting in.

[0132] When avoiding a target that has passed behind the vehicle and then slowed down, or when avoiding a target that is running alongside the vehicle, the vehicle's driving scenario is one where the longitudinal collision distance is relatively small when the target appears.

[0133] This optional implementation can prevent the avoidance function from being triggered in scenarios where the longitudinal collision distance is small by identifying such scenarios. This avoids excessive changes in vehicle speed caused by triggering the avoidance function in such scenarios, thereby preventing excessive changes in vehicle speed from increasing driving risks and further improving driving safety.

[0134] In the above embodiments, one implementation method for determining whether the target has overtaken the vehicle from behind and then decelerated based on the target's speed and position is as follows:

[0135] Based on the position of the target to be avoided, determine the longitudinal coordinate of the target at the current moment, compare the longitudinal coordinate of the target at the current moment with the longitudinal coordinate of the vehicle at the current moment, and determine whether the target to be avoided is in front of the vehicle at the current moment based on the comparison result.

[0136] Based on the position of the target to be avoided, obtain the longitudinal coordinate of the target at the previous moment, compare the longitudinal coordinate of the target at the previous moment with the longitudinal coordinate of the vehicle at the previous moment, and determine whether the target to be avoided was behind the vehicle at the previous moment based on the comparison result.

[0137] If the target being avoided is in front of the vehicle at the current moment, and the target being avoided was behind the vehicle at the previous moment, then it is determined that the target being avoided will overtake the vehicle from behind.

[0138] Obtain the current speed of the vehicle and the next speed of the vehicle to avoid the target;

[0139] When the next moment arrives, the current speed of the vehicle being avoided is compared with the speed of the vehicle being avoided at the next moment. If the speed of the vehicle being avoided at the next moment is less than the current speed of the vehicle being avoided, it is determined that the vehicle being avoided is in a deceleration state. Finally, it is determined that the vehicle being avoided will first overtake the vehicle from behind and then decelerate.

[0140] In the above embodiments, one approach to determine whether the target to be avoided is parallel to the vehicle during the cut-in is as follows:

[0141] Based on the position of the target to be avoided, determine the longitudinal coordinate of the target at the current moment, compare the longitudinal coordinate of the target at the current moment with the longitudinal coordinate of the vehicle at the current moment, and determine whether the target to be avoided is parallel to the vehicle at the current moment based on the comparison result.

[0142] Based on the position of the target to be avoided, obtain the longitudinal coordinate of the target at the previous moment, compare the longitudinal coordinate of the target at the previous moment with the longitudinal coordinate of the vehicle at the previous moment, and determine whether the target to be avoided was behind the vehicle at the previous moment based on the comparison result.

[0143] If the target being avoided is parallel to the vehicle at the current moment, and the target being avoided was behind the vehicle at the previous moment, then it is determined that the target being avoided is cutting in from behind the vehicle.

[0144] Obtain the current speed of the target being avoided;

[0145] If the speed of the target being avoided is the same as the speed of the vehicle, then the target being avoided is determined to be parallel to the vehicle when it cuts in.

[0146] In the above implementation, "avoiding the target while running parallel to the vehicle at the current moment" can mean that part of the target's body has passed the vehicle at the current moment, but has not completely passed it, and its speed is the same as the vehicle's.

[0147] In certain specific scenarios, if a vehicle is currently changing lanes, triggering the evasive maneuver function can compromise driving safety. For example, if the vehicle is changing lanes to the left and the evasive maneuver function is triggered, it might take emergency braking or turn the steering wheel to the right, causing the driver to lose control and change lanes to the left. This contradicts the actual necessary action and could lead to a traffic accident. Furthermore, after changing lanes, the evasive maneuver target should be the target after the lane change. Using the target before or after the lane change as the evasive maneuver target may cause the evasive maneuver function to take incorrect actions.

[0148] To address this scenario, in an optional implementation, the vehicle changing lanes is considered as one of the target scenarios to avoid triggering the avoidance function during lane changes. Accordingly, the method in this embodiment further includes the following steps:

[0149] Determine whether a vehicle is changing lanes based on its location and lane information;

[0150] When a vehicle is changing lanes, the driving scenario is called a lane-changing scenario.

[0151] Therefore, this optional implementation method can prevent the avoidance function from being triggered in the lane-changing scenario by recognizing the lane-changing scenario of the vehicle, thereby further improving driving safety.

[0152] In the above embodiments, one approach to determining whether a vehicle is changing lanes based on its position and lane information is as follows:

[0153] The vehicle's current lane number and surrounding lane numbers are determined based on lane information.

[0154] Based on the vehicle's position, determine whether the vehicle is changing lanes from its current lane number to the lane corresponding to an adjacent lane number. If so, determine that the vehicle is in the process of changing lanes.

[0155] In addition, the vehicle's posture is unstable immediately after it changes lanes. Triggering the avoidance function when the vehicle's posture is unstable will also reduce the vehicle's driving safety.

[0156] In this scenario, the embodiment of this application considers the vehicle's driving scenario as a lane-changing scenario within a few seconds after the vehicle completes the lane change.

[0157] In some specific scenarios, although there may be surrounding moving targets within a preset time range for longitudinal collision with the vehicle, these targets are in different lanes and do not show any tendency to approach the vehicle laterally. In such cases, triggering the avoidance function is unnecessary. Triggering the avoidance function would lower its triggering conditions, making it easier to activate and leading to frequent triggering. Furthermore, generally, the avoidance function performs one action for one target; if the target changes frequently, the avoidance function may be triggered more often.

[0158] For the above scenario, the vehicle is controlled to avoid the target based on its state information and surrounding environment information, including the following sub-steps:

[0159] The lateral distance between the vehicle and the target to be avoided is determined based on the vehicle's position and the target's position.

[0160] Determine the lateral velocity of the target relative to the ground based on the target's velocity.

[0161] Obtain the distance limit values ​​between the vehicle and the left and right lane lines;

[0162] When the longitudinal collision distance between the vehicle and the surrounding moving target is within a preset time distance value range, and the absolute value of the lateral distance between the avoidance target and the vehicle is within a preset distance value range, and the lateral velocity of the avoidance target relative to the ground is within a preset speed value range, and the avoidance target does not change within the system cycle, an avoidance activation command is determined and an avoidance offset is determined based on the lane width and the distance limit values ​​between the vehicle and the left and right lane lines.

[0163] Vehicles are controlled to avoid targets based on avoidance offset.

[0164] Therefore, the aforementioned optional implementation method triggers the avoidance function only when the longitudinal collision distance between the vehicle and the surrounding moving target is within a preset time range, the absolute value of the lateral distance between the avoidance target and the vehicle is within a preset distance range, the lateral velocity of the avoidance target relative to the ground is within a preset speed range, and the avoidance target does not change within the system cycle. The avoidance offset is then determined based on the lane width and the distance limits between the vehicle and the left and right lane lines. When these conditions are not met, the avoidance function is not triggered, thus avoiding repeated advance and retreat avoidance maneuvers and further improving comfort.

[0165] In the above embodiments, the lateral distance between the avoidance target and the vehicle refers to the distance between the avoidance target and the vehicle in the lateral coordinate axis direction, with the vehicle as the reference target. Further, the preset distance value range can be between 0.5 meters and 1 meter. For example, if the avoidance target is to the left of the vehicle and the lateral distance between them is 0.6, then the absolute value of the lateral distance between the avoidance target and the vehicle is considered to be within the preset distance value range.

[0166] In the above embodiments, the lateral velocity of the target relative to the ground refers to the lateral velocity of the target relative to the ground, with the ground as the reference target. Further, the preset velocity range can be above zero. If the lateral velocity of the target relative to the ground is in a value range greater than zero, it is considered that the target has no tendency to approach the vehicle.

[0167] In the above implementation, the distance limit values ​​for the left and right lane lines include the distance limit value for the left lane line and the distance limit value for the right lane line. If changing lanes to the left, the deviation is calculated using the distance limit value for the left lane line and the width of the lane. If changing lanes to the right, the deviation is calculated using the distance limit value for the right lane line and the width of the lane.

[0168] In the above implementation, the deviation amount refers to the amount by which the vehicle needs to change position to the left or right based on its current position.

[0169] In the above embodiments, the fact that the avoidance target does not change within the system cycle may mean that the ID of the avoidance target does not change within the system cycle. The system cycle may be 1 second or other values. This application embodiment does not limit this.

[0170] In some scenarios, the existing avoidance function is limited to large trucks as avoidance targets, and cannot avoid small two-wheeled vehicles, thus limiting the scope of application of the avoidance function.

[0171] In an alternative implementation, the target to be avoided includes a two-wheeled vehicle.

[0172] This optional implementation can target two-wheeled vehicles, thus increasing the range of the obstacle avoidance function.

[0173] In the above optional embodiments, two-wheeled vehicles include small vehicles such as bicycles and motorized motorcycles.

[0174] Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of a vehicle avoidance device provided in an embodiment of this application. Figure 2 As shown, the vehicle avoidance device in this application embodiment includes the following functional modules:

[0175] The acquisition module 201 is used to acquire vehicle status information and surrounding environment information;

[0176] The determination module 202 is used to determine the avoidance target based on the vehicle's state information and surrounding environment information;

[0177] The control module 203 is used to control the vehicle to exit the avoidance mode when the vehicle's driving scenario is the target scenario; otherwise, it controls the vehicle to avoid the target based on the vehicle's status information and surrounding environment information. The target scenario includes the scenario where the vehicle is flanked by both sides.

[0178] When the device in this embodiment of the application identifies a target to avoid, it can avoid that target. Simultaneously, by controlling the vehicle to exit the avoidance mode in a scenario where it is surrounded from both sides, it can prevent the vehicle from frequently avoiding targets located on the left and right sides of the vehicle in such a scenario, thus preventing the vehicle from swaying left and right and reducing its driving safety, thereby improving driving safety.

[0179] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. For example... Figure 3 As shown, the electronic device in this application embodiment includes

[0180] Processor 301; and

[0181] The memory 302 is configured to store machine-readable instructions that, when executed by a processor, perform a vehicle avoidance method as described in any of the foregoing embodiments.

[0182] The electronic device in this application embodiment, by executing a vehicle avoidance method, can avoid a target when it is detected. Simultaneously, by controlling the vehicle to exit the avoidance mode in a scenario where it is surrounded from both sides, it can prevent the vehicle from frequently avoiding targets located on the left and right sides of the vehicle in such a scenario, thus avoiding lateral swaying and reducing vehicle driving safety, thereby improving vehicle driving safety.

[0183] This application also provides a storage medium storing a computer program, which is executed by a processor as a vehicle avoidance method according to any of the foregoing embodiments.

[0184] The storage medium in this application embodiment, by executing a vehicle avoidance method, can avoid a target when it is identified. Simultaneously, by controlling the vehicle to exit avoidance mode in a scenario where it is flanked from both sides, it can prevent the vehicle from frequently avoiding targets located on the left and right sides of the vehicle in such a scenario, thus avoiding lateral swaying and improving vehicle driving safety.

[0185] As an example, with Figure 4 Taking the target perception diagram shown as an example, the vehicle perceives four targets: TG1, TG2, TG3, and TG4. The longitudinal TTC is used as an important condition for filtering and avoiding targets. The longitudinal TTC is the longitudinal collision time distance between the vehicle and the target, and the formula is "the relative longitudinal distance between the vehicle and the target / (vehicle speed - target speed)". The magnitude and sign of the longitudinal TTC can intuitively show the collision trend between the vehicle and the target. A smaller longitudinal TTC means that the vehicle is about to collide with the target. At the same time, a positive longitudinal TTC means that the target is approaching the vehicle, while a negative value means that the target is moving away from the vehicle.

[0186] Furthermore, the system uses visual sensors to determine whether the surrounding lane environment meets the avoidance conditions. First, the width of the lane where the vehicle is located cannot be zero, the confidence level of the lane lines on both sides needs to meet specific conditions, and the vehicle speed cannot be too low. A certain speed condition needs to be met to meet the activation conditions of the intelligent avoidance function.

[0187] Further, in the target selection phase, the types of targets to be avoided include trucks and two-wheeled vehicles, namely motorcycles, electric bicycles, and bicycles. Therefore, the type of vehicle to be avoided must meet the above-mentioned vehicle types to be considered as a target. Subsequently, it is necessary to determine whether the longitudinal TTC of each vehicle target meets the conditions. Intelligent avoidance selects the target vehicle with a positive longitudinal TTC and the smallest longitudinal TTC value as the target vehicle. Therefore, vehicles that overtake the vehicle from behind will not be selected. At the same time, there is also a target locking logic in the target selection phase. When the avoidance function is activated, if the target to be avoided is close to the vehicle and the target's longitudinal TTC is negative, intelligent avoidance will lock the target as the avoidance target and maintain the current avoidance function until the target disappears or the longitudinal distance of the target becomes too far.

[0188] Furthermore, after target selection, intelligent obstacle avoidance needs to determine whether any special scenarios exist. The existence of such special scenarios indicates that the vehicle is not suitable for avoidance maneuvers under the current conditions. Therefore, if special scenarios exist, the intelligent obstacle avoidance function will not be activated. These special scenarios are the target scenarios, including:

[0189] In a scenario where vehicles are sandwiched between two sides, meaning there are vehicles that meet the avoidance criteria on both sides of the vehicle and the longitudinal distance between the targets on the left and right sides is relatively short, the vehicle will not activate the intelligent avoidance function.

[0190] In scenarios involving small-radius curves, where the vehicle is on a curve with a small radius and there are vehicles on the left or right sides that meet the avoidance criteria, the vehicle will not activate the intelligent avoidance function.

[0191] In a scenario where the target is about to cut out of the lane, the system determines whether the vehicle or target has a tendency to cut out of the lane based on the lateral distance between the target and the vehicle, as well as the lateral velocity of the target relative to the ground. If a cutting-out tendency is determined, the vehicle will not activate the intelligent lane avoidance function.

[0192] In scenarios where longitudinal TTC is small when a target appears, such as when the vehicle decelerates after passing behind the vehicle and moving a few meters in front, or when a target cuts in and runs parallel to the vehicle, the vehicle will not activate the intelligent avoidance function.

[0193] In the scenario of a vehicle changing lanes, that is, during the process of changing lanes and within a few seconds after the lane change, the overall posture of the vehicle is unstable and it is not suitable to perform lateral evasive maneuvers. Therefore, the intelligent avoidance function will not be activated when this scenario occurs.

[0194] If the aforementioned special scenarios do not occur during the intelligent obstacle avoidance function process, the system will enter the intelligent obstacle avoidance activation phase. This phase involves conditional determinations of the longitudinal TTC of the obstacle, the lateral distance between the obstacle and the vehicle, the lateral speed of the obstacle relative to the ground, and the obstacle ID. Activating the obstacle avoidance function requires ensuring that the longitudinal TTC is within a set range, the absolute value of the lateral distance between the obstacle and the vehicle is not excessive, the lateral distance between the obstacle and the ground is within a set range and cannot be excessive, and the obstacle ID remains consistent throughout a single system cycle—meaning the obstacle cannot change during intelligent obstacle avoidance activation. A minimum longitudinal distance between the obstacle and the vehicle is also set to ensure a reasonable lower limit for the longitudinal distance between the vehicle and the obstacle when the function is activated. It is important to emphasize that the longitudinal TTC range changes with the relative speed between the obstacle and the vehicle. Compared to the mainstream fixed-range solutions on the market, the variable range used in this example also prevents excessive longitudinal distance between the vehicle and the obstacle during intelligent obstacle avoidance activation, which could lead to a poor driver experience.

[0195] Once the intelligent obstacle avoidance function is activated, it's necessary to consider the conditions under which it will exit and how to calculate the duration of the exit. The main conditions for exiting the obstacle avoidance function include: the target disappearing, the absolute value of the lateral distance between the target and the vehicle being too large, and the target continuously moving away from the vehicle. If any of these conditions exist, the intelligent obstacle avoidance exit phase must be initiated. When the intelligent obstacle avoidance enters the exit phase, this patent uses a fixed delay scheme to determine the duration of the exit. That is, when the above exit conditions are met, the system maintains the obstacle avoidance function for a certain period. If no new vehicle target meeting the obstacle avoidance conditions appears during this period, the intelligent obstacle avoidance function will immediately exit, canceling its activation. Thus, once the intelligent obstacle avoidance function officially exits, the entire intelligent obstacle avoidance function process is completed.

[0196] During the activation of the intelligent obstacle avoidance function, in addition to sending a request to activate the function, the system also needs to output an offset to the control terminal. This allows the control terminal to control the vehicle to perform the corresponding left or right obstacle avoidance maneuver based on the offset. In this example, the offset output is limited by the lane width and the distance between the vehicle and the left and right lane lines to avoid outputting an excessively large offset when the lane is narrow, which could lead to the vehicle crossing the lane lines during the avoidance process. The offset output is limited to the maximum value based on the lane width and the distance between the vehicle and the left and right lane lines, and the offset changes linearly based on these conditions. This allows intelligent obstacle avoidance to adapt to avoidance operations under different lane conditions in real time, preventing the vehicle from crossing the lane lines or getting too close to them. Simultaneously, based on the position of the obstacle relative to the vehicle, the offset is directionally defined: a negative offset indicates the vehicle will perform an avoidance maneuver to the left, and a positive offset indicates the vehicle will perform an avoidance maneuver to the right.

[0197] During the intelligent avoidance activation period, the intelligent avoidance system outputs an avoidance function activation request and the corresponding avoidance action offset to the control terminal. The control terminal uses the offset to perform an avoidance action in the corresponding direction, thus realizing intelligent avoidance action for a specific target.

[0198] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interface; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0199] Furthermore, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0200] Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0201] It should be noted that if a function is implemented as a software module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0202] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0203] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for avoiding a vehicle, characterized in that, The method includes: Obtain vehicle status information and surrounding environment information; The avoidance target is determined based on the vehicle's status information and the surrounding environment information; When the driving scenario of the vehicle is the target scenario, the vehicle is controlled to exit the avoidance mode; otherwise, based on the vehicle's status information and the surrounding environment information, the vehicle is controlled to avoid the avoidance target. The target scenario includes a scenario where the vehicle is flanked by both sides. The vehicle's surrounding environment information includes the vehicle speed and position of surrounding moving targets, and the vehicle's status information includes the vehicle speed and position. And, the step of determining the avoidance target based on the vehicle's state information and the surrounding environment information includes: The longitudinal distance of the vehicle relative to the surrounding moving targets is obtained based on the position of the vehicle and the position of the surrounding moving targets; Based on the longitudinal distance of the vehicle relative to the surrounding moving target, the speed of the vehicle, and the speed of the surrounding moving target, the longitudinal collision time distance between the vehicle and the surrounding moving target is obtained; When the longitudinal collision distance is positive, the surrounding moving target is identified as the avoidance target; The surrounding environment information also includes lane information; Furthermore, before determining the avoidance target based on the vehicle's state information and the surrounding environment information, the method further includes: The width of the lane in which the vehicle is located and the confidence level of the lanes surrounding the vehicle are determined based on the lane information. When the width of the lane where the vehicle is located is not zero, and the confidence level of the lanes surrounding the vehicle is greater than or equal to a preset confidence threshold, determine whether the vehicle speed is greater than or equal to a preset speed threshold. When the vehicle's speed is greater than or equal to a preset speed threshold, the process of determining the avoidance target based on the vehicle's state information and the surrounding environment information is triggered.

2. The method as described in claim 1, characterized in that, After determining the avoidance target based on the vehicle's state information and the surrounding environment information, the method further includes: Based on the position of the avoidance target and the position of the vehicle, determine whether the avoidance target appears on both sides of the vehicle; When the avoidance target appears on both sides of the vehicle, it is determined whether the longitudinal distance between the vehicle and each avoidance target is less than or equal to a preset longitudinal distance threshold. When the longitudinal distance between the vehicle and each of the avoidance targets is less than or equal to the preset longitudinal distance threshold, the driving scenario of the vehicle is a scenario where the vehicle is flanked by both sides.

3. The method as described in claim 2, characterized in that, The target scenario also includes small-radius curve scenarios; Furthermore, the method further includes: When the avoidance target appears on both sides of the vehicle, it is determined whether the lane where the vehicle is located is a curve based on the lane information, and whether the radius of the curve is less than or equal to a preset radius threshold. If the lane where the vehicle is located is the curve, and the radius of the curve is less than or equal to the preset radius threshold, then the vehicle's driving scenario is the small-radius curve scenario.

4. The method as described in claim 1, characterized in that, The target scenario also includes scenarios where the longitudinal collision distance is small when the target appears; Furthermore, the method further includes: Based on the speed of the target being avoided and the position of the target being avoided, it is determined whether the target being avoided slows down after passing behind the vehicle to the front of the vehicle, or whether the target being avoided runs parallel to the vehicle when cutting in. When the vehicle decelerates after the avoidance target passes behind the vehicle and moves in front of the vehicle, or when the avoidance target cuts in parallel with the vehicle, the driving scenario of the vehicle is a scenario where the longitudinal collision distance is small when the target appears.

5. The method as described in claim 1, characterized in that, The target scenarios also include scenarios where vehicles change lanes; Furthermore, the method further includes: Based on the vehicle's position and the lane information, determine whether the vehicle is in the process of changing lanes; When the vehicle is in the process of changing lanes, the driving scenario of the vehicle is a lane-changing scenario.

6. The method as described in claim 1, characterized in that, The method of controlling the vehicle to avoid the target based on the vehicle's state information and the surrounding environment information includes: The lateral distance of the avoidance target relative to the vehicle is determined based on the position of the vehicle and the position of the avoidance target; The lateral velocity of the target relative to the ground is determined based on the velocity of the target being avoided; Obtain the distance limit values ​​between the vehicle and the left and right lane lines; When the longitudinal collision distance between the vehicle and the surrounding moving target is within a preset time distance value range, and the absolute value of the lateral distance between the avoidance target and the vehicle is within a preset distance value range, and the lateral velocity of the avoidance target relative to the ground is within a preset speed value range, and the avoidance target does not change within the system cycle, an avoidance activation command is determined and an avoidance offset is determined based on the width of the lane and the distance limit values ​​between the vehicle and the left and right lane lines; The vehicle is controlled to avoid the target based on the avoidance offset.

7. The method as described in claim 1, characterized in that, The targets to be avoided include two-wheeled vehicles.

8. A vehicle avoidance device, characterized in that, The device includes: The acquisition module is used to acquire vehicle status information and surrounding environment information; The determination module is used to determine the avoidance target based on the vehicle's state information and the surrounding environment information; The control module is used to control the vehicle to exit the avoidance mode when the driving scenario of the vehicle is the target scenario; otherwise, it controls the vehicle to avoid the avoidance target based on the vehicle's status information and the surrounding environment information, wherein the target scenario includes a scenario where the vehicle is surrounded by both sides. The vehicle's surrounding environment information includes the speed of surrounding moving targets and the position of the surrounding moving targets, and the vehicle's state information includes the vehicle's speed and the vehicle's position. And, the step of determining the avoidance target based on the vehicle's state information and the surrounding environment information includes: The longitudinal distance of the vehicle relative to the surrounding moving targets is obtained based on the position of the vehicle and the position of the surrounding moving targets; Based on the longitudinal distance of the vehicle relative to the surrounding moving target, the speed of the vehicle, and the speed of the surrounding moving target, the longitudinal collision time distance between the vehicle and the surrounding moving target is obtained; When the longitudinal collision distance is positive, the surrounding moving target is identified as the avoidance target; The surrounding environment information also includes lane information; Furthermore, before determining the avoidance target based on the vehicle's state information and the surrounding environment information, the method further includes: The width of the lane in which the vehicle is located and the confidence level of the lanes surrounding the vehicle are determined based on the lane information. When the width of the lane where the vehicle is located is not zero, and the confidence level of the lanes surrounding the vehicle is greater than or equal to a preset confidence threshold, determine whether the vehicle speed is greater than or equal to a preset speed threshold. When the vehicle's speed is greater than or equal to a preset speed threshold, the process of determining the avoidance target based on the vehicle's state information and the surrounding environment information is triggered.

9. An electronic device, characterized in that, include: processor; as well as A memory configured to store machine-readable instructions that, when executed by the processor, perform the vehicle avoidance method as described in any one of claims 1-7.

10. A storage medium, characterized in that, The storage medium stores a computer program, which is executed by a processor as the vehicle avoidance method as described in any one of claims 1-7.

Images