Vehicle avoidance method and vehicle

Abstract

The invention relates to a vehicle avoiding method and a vehicle. The method comprises the steps that under the condition that it is detected that a second vehicle exists in the peripheral range of a first vehicle, first driving information of the first vehicle and second driving information of the second vehicle are obtained, and the passing priority of the second vehicle is higher than that of the first vehicle; under the condition that it is determined that the first vehicle needs to be controlled to avoid the second vehicle according to the first driving information and the second driving information, detecting an available avoidance space in the peripheral range of the vehicle; if the available avoidance space meets the avoidance condition of the first vehicle, the first vehicle is controlled to conduct lane changing avoidance on the second vehicle; and if the available avoidance space does not meet the avoidance condition, the first vehicle is controlled to perform inclined avoidance on the second vehicle. By adopting the method, the vehicle can be accurately avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent driving, in particular to a vehicle avoiding method and a vehicle. BACKGROUND

[0002] With the development of intelligent driving technology, how to avoid traffic conflict problems in the process of intelligent driving has become a hot topic in the field of intelligent driving. By controlling a vehicle to avoid another vehicle with higher traffic priority, such as a special vehicle, traffic conflicts can be avoided and accident risks can be reduced.

[0003] In the traditional technology, the method for avoiding a vehicle generally sets a visual sensor or an auditory sensor for the ego vehicle in advance. The ego vehicle identifies a vehicle with higher traffic priority through the visual sensor or the auditory sensor, and then locates the vehicle with higher traffic priority. According to the positioning result of the identified vehicle, the ego vehicle is controlled to avoid the vehicle.

[0004] However, the current vehicle avoiding method has the problem of inaccuracy. SUMMARY

[0005] Therefore, it is necessary to provide an accurate vehicle avoiding method, a vehicle, a computer device, a computer readable storage medium and a computer program product in view of the above technical problems.

[0006] In a first aspect, the present application provides a vehicle avoiding method, comprising:

[0007] In a case where it is detected that there is a second vehicle in the vehicle surrounding range of a first vehicle, first driving information of the first vehicle and second driving information of the second vehicle are obtained, wherein the traffic priority of the second vehicle is higher than that of the first vehicle;

[0008] In a case where it is determined according to the first driving information and the second driving information that the first vehicle needs to be controlled to avoid the second vehicle, available avoiding space in the vehicle surrounding range is detected;

[0009] If the available avoiding space meets the avoiding condition of the first vehicle, the first vehicle is controlled to avoid the second vehicle by lane changing;

[0010] If the available avoiding space does not meet the avoiding condition, the first vehicle is controlled to avoid the second vehicle by tilting.

[0011] In a second aspect, the present application further provides a vehicle, which comprises a vehicle avoiding device, and the device comprises:

[0012] The avoiding vehicle information acquisition module is configured to acquire first driving information of the first vehicle and second driving information of the second vehicle in a case where it is detected that the second vehicle exists in the vehicle surrounding range of the first vehicle, and the passing priority of the second vehicle is higher than the passing priority of the first vehicle;

[0013] The avoiding judgment module is configured to detect available avoiding space in the vehicle surrounding range in a case where it is determined according to the first driving information and the second driving information that the first vehicle needs to be controlled to avoid the second vehicle;

[0014] The first avoiding module is configured to control the first vehicle to perform lane-changing avoiding on the second vehicle if the available avoiding space meets avoiding conditions of the first vehicle.

[0015] The second avoiding module is configured to control the first vehicle to perform tilting avoiding on the second vehicle if the available avoiding space does not meet the avoiding conditions.

[0016] In a third aspect, the present application further provides a computer device, comprising a memory and a processor, the memory stores a computer program, and the processor realizes the following steps when executing the computer program:

[0017] In a case where it is detected that the second vehicle exists in the vehicle surrounding range of the first vehicle, first driving information of the first vehicle and second driving information of the second vehicle are acquired, and the passing priority of the second vehicle is higher than the passing priority of the first vehicle;

[0018] In a case where it is determined according to the first driving information and the second driving information that the first vehicle needs to be controlled to avoid the second vehicle, available avoiding space in the vehicle surrounding range is detected;

[0019] If the available avoiding space meets avoiding conditions of the first vehicle, the first vehicle is controlled to perform lane-changing avoiding on the second vehicle.

[0020] If the available avoiding space does not meet the avoiding conditions, the first vehicle is controlled to perform tilting avoiding on the second vehicle.

[0021] In a fourth aspect, the present application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the following steps:

[0022] In a case where it is detected that the second vehicle exists in the vehicle surrounding range of the first vehicle, first driving information of the first vehicle and second driving information of the second vehicle are acquired, and the passing priority of the second vehicle is higher than the passing priority of the first vehicle;

[0023] In a case where it is determined according to the first driving information and the second driving information that the first vehicle needs to be controlled to avoid the second vehicle, available avoiding space in the vehicle surrounding range is detected;

[0024] if the available avoidance space meets the avoidance condition of the first vehicle, controlling the first vehicle to perform lane-changing avoidance on the second vehicle;

[0025] if the available avoidance space does not meet the avoidance condition, controlling the first vehicle to perform inclined avoidance on the second vehicle.

[0026] In the fifth aspect, the present application also provides a computer program product, comprising a computer program which, when executed by a processor, implements the following steps:

[0027] In the case where it is detected that the second vehicle exists in the vehicle surrounding range of the first vehicle, obtaining first driving information of the first vehicle and second driving information of the second vehicle, wherein the passing priority of the second vehicle is higher than that of the first vehicle;

[0028] In the case where it is determined according to the first driving information and the second driving information that the first vehicle needs to perform avoidance on the second vehicle, detecting available avoidance space in the vehicle surrounding range;

[0029] if the available avoidance space meets the avoidance condition of the first vehicle, controlling the first vehicle to perform lane-changing avoidance on the second vehicle;

[0030] if the available avoidance space does not meet the avoidance condition, controlling the first vehicle to perform inclined avoidance on the second vehicle.

[0031] The vehicle avoidance method, vehicle, computer device, computer readable storage medium and computer program product accurately determine whether the first vehicle needs to perform avoidance on the second vehicle according to the first driving information of the first vehicle and the second driving information of the second vehicle in the case where it is detected that the second vehicle exists in the vehicle surrounding range of the first vehicle, and detect available avoidance space in the vehicle surrounding range in the case where the first vehicle needs to perform avoidance on the second vehicle. If the available avoidance space meets the avoidance condition of the first vehicle, the first vehicle performs lane-changing avoidance on the second vehicle. If the available avoidance space does not meet the avoidance condition, the first vehicle performs inclined avoidance on the second vehicle. In the whole process, the first driving information of the first vehicle, the second driving information of the second vehicle and the available avoidance space in the vehicle surrounding range are used to determine whether to perform lane-changing avoidance or inclined avoidance on the second vehicle from the driving state of the first vehicle and the second vehicle and the surrounding environment state of the first vehicle. In different cases, the second vehicle with a higher passing priority can be provided with the maximum passing space. BRIEF DESCRIPTION OF DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the following will briefly introduce the drawings needed to be used in the description of the embodiments of the present application or the related art. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0033] Figure 1 An application environment diagram of the vehicle avoiding method in one embodiment;

[0034] Figure 2 A flowchart of the vehicle avoiding method in one embodiment;

[0035] Figure 3 A flowchart of the vehicle avoiding method in another embodiment;

[0036] Figure 4 A flowchart of the vehicle avoiding method in another embodiment;

[0037] Figure 5 A flowchart of the vehicle avoiding method in one specific application embodiment;

[0038] Figure 6 A structural block diagram of the vehicle avoiding device in one embodiment;

[0039] Figure 7 An internal structure diagram of the computer device in one embodiment. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of the present application more clear, the following will further describe the present application in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are used to explain the present application, and are not used to limit the present application.

[0041] In the conventional technology, the method for avoiding vehicles generally sets a visual sensor or an auditory sensor for the ego vehicle in advance, identifies a vehicle with a higher traffic priority through the visual sensor or the auditory sensor of the ego vehicle, positions the vehicle with a higher traffic priority, and then controls the ego vehicle to avoid the vehicle with a higher traffic priority, such as changing lanes to another lane, according to the positioning result of the identified vehicle. In actual application, the vehicle with a higher traffic priority includes but is not limited to special vehicles such as police cars, ambulances and military vehicles. In addition, the vehicle with a higher traffic priority can also be a plurality of continuous vehicles with a higher traffic priority, that is, a special vehicle team.

[0042] However, the current vehicle avoiding method lacks a certain accuracy in the avoiding process, and cannot accurately avoid the vehicle with a higher traffic priority according to the identification result.

[0043] To this end, the application provides a more accurate vehicle avoidance method. In the case where a second vehicle with a higher traffic priority is detected in the vehicle range of a first vehicle, whether the first vehicle needs to avoid the second vehicle is accurately determined according to first driving information of the first vehicle and second driving information of the second vehicle, and in the case where the first vehicle needs to avoid the second vehicle, available avoidance space in the vehicle range is detected. If the available avoidance space meets avoidance conditions of the first vehicle, the first vehicle is controlled to avoid the second vehicle by changing lanes. If the available avoidance space does not meet the avoidance conditions, the first vehicle is controlled to avoid the second vehicle by tilting. In the whole process, the first driving information of the first vehicle, the second driving information of the second vehicle, and the available avoidance space in the vehicle range are used to accurately determine whether the second vehicle is avoided by changing lanes or by tilting, and the maximum traffic space is provided for the second vehicle with a higher traffic priority in different cases.

[0044] The vehicle avoidance method provided by the application can be applied to an application environment as shown in Figure 1 The preset space range 1000 is provided with a plurality of drivable lanes, including lane 1,..., lane n. A plurality of vehicles 102 are driven on each lane, and each vehicle 102 is provided with a vehicle controller 104 of the vehicle. Each vehicle controller 104 is in communication connection with a cloud 106. The data storage system can be integrated on the cloud 106 or placed on the vehicle controller 104 of each vehicle 102.

[0045] Specifically, all vehicles 102 drive in the preset space range as shown in Figure 1 The vehicle controller 104 of each vehicle 102 detects whether a second vehicle with a higher traffic priority exists in the vehicle range of the vehicle in real time, and the vehicle is taken as a first vehicle. At this time, the vehicle controller 104 of the first vehicle detects the second vehicle in the vehicle range of the first vehicle, obtains first driving information of the first vehicle and second driving information of the second vehicle, and determines whether the first vehicle needs to avoid the second vehicle according to the first driving information and the second driving information. In the case where the first vehicle needs to avoid the second vehicle, the available avoidance space in the vehicle range is detected. If the available avoidance space meets the avoidance conditions of the first vehicle, the first vehicle is controlled to avoid the second vehicle by changing lanes. If the available avoidance space does not meet the avoidance conditions, the first vehicle is controlled to avoid the second vehicle by tilting.

[0046] The cloud 106 can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0047] In one exemplary embodiment, as shown in Figure 2 , a vehicle avoidance method is provided, which is applied to the vehicle controller 104 of the first vehicle in Figure 1 , including the following S200 to S800. Among them:

[0048] S200, in the case of detecting the existence of the second vehicle in the vehicle range of the first vehicle, obtaining the first driving information of the first vehicle and the second driving information of the second vehicle.

[0049] Among them, the passing priority of the second vehicle is higher than that of the first vehicle. The vehicle range of the vehicle refers to the three-dimensional space area covering the periphery of the vehicle body which is directly related to the driving safety and environmental interaction, and the size of the three-dimensional space area can be set according to the actual avoidance demand.

[0050] Specifically, all vehicles drive in the preset space range as shown in Figure 1 , and the vehicle controller of each vehicle detects whether there is a second vehicle with a higher passing priority in the vehicle range of the vehicle in real time, and takes the vehicle as the first vehicle, and the passing priority of the first vehicle is lower than that of the second vehicle, so the first vehicle needs to avoid the second vehicle to make the second vehicle drive first.

[0051] In the case that the vehicle controller of the first vehicle detects the existence of the second vehicle in the vehicle range of the first vehicle, the first driving information of the vehicle and the second driving information of the second vehicle will be obtained, and the driving information includes but is not limited to the speed, lane, position and other information of the vehicle.

[0052] In one embodiment, each vehicle is provided with a speed sensor, a position sensor and the like, so that when detecting the first driving information of the vehicle, the speed, lane, position, orientation and the like of the first vehicle can be detected through the multiple sensors provided, in addition, the positioning information of the vehicle can also be obtained through the GPS (Global Positioning System, Global Positioning System) or Beidou positioning module and other wireless communication tools, and the specific detection method is not limited here.

[0053] When detecting the second driving information of the second vehicle, the second driving information of the second vehicle can be detected through visual recognition means, and / or the second driving information of the second vehicle can be detected through auditory recognition means. In practical application, the visual recognition means includes at least one of visual sensor, laser sensor and the like, and the auditory recognition means includes at least one of audio recognition, voice prompt message recognition and the like of the second vehicle, and the specific detection method is not limited here.

[0054] Further, according to the first driving information of the first vehicle and the second driving information of the second vehicle, the relative position information between the first vehicle and the second vehicle can also be obtained, which includes the relative lane information and the relative position information between the first vehicle and the second vehicle, for example, the relative lane information includes whether the special vehicle and the ego vehicle are in the same lane, and the relative position information includes whether the special vehicle is in the front area of the driving route of the ego vehicle, etc.

[0055] In one embodiment, when the driving information includes lane information, the vehicle lane judgment needs to be continuously performed, and when the lane information does not change for a period of time, the lane information is taken as the actual driving information.

[0056] S400, according to the first driving information and the second driving information, detecting the available avoidance space in the vehicle range of the first vehicle.

[0057] The available avoidance space in the vehicle range of the first vehicle refers to the space available for avoidance when the first vehicle avoids the second vehicle, and here refers to the available avoidance space on the adjacent lane of the lane where the first vehicle is located. The available avoidance space needs to meet the avoidance condition of the first vehicle to control the first vehicle to avoid.

[0058] Specifically, according to the first driving information and the second driving information, it is judged whether the first vehicle needs to be controlled to avoid the second vehicle, if the first vehicle does not need to be controlled to avoid the second vehicle, the first vehicle normally drives; if the first vehicle needs to be controlled to avoid the second vehicle, in order to maximize the passing space of the second vehicle, the lane changing avoidance mode is generally preferred, therefore, at this time, the first vehicle cannot be directly controlled to avoid the second vehicle, but the available avoidance space in the vehicle range of the first vehicle needs to be detected first, and it is judged whether the available avoidance space meets the avoidance condition of the first vehicle, when the available avoidance space meets the avoidance condition of the first vehicle, the first vehicle is controlled to change lane to avoid the second vehicle.

[0059] In actual application, the available avoidance space in the vehicle range of the first vehicle can be realized by configuring visual, millimeter wave, laser radar and other sensors to perform space perception on the lane changing position of the first vehicle.

[0060] S600, if the available avoidance space meets the avoidance condition of the first vehicle, the first vehicle is controlled to change lane to avoid the second vehicle.

[0061] The available avoidance space needs to meet the avoidance condition of the first vehicle, so as to control the first vehicle to avoid. The avoidance condition of the first vehicle can actually be that the available avoidance space needs to meet the avoidance required space of the first vehicle, wherein the avoidance required space of the first vehicle is determined by the size of the first vehicle itself.

[0062] Specifically, if the available avoidance space meets the avoidance condition of the first vehicle, it means that there is available avoidance space for the first vehicle to avoid by lane changing, and the first vehicle can avoid the second vehicle by lane changing at this time. At this time, the vehicle controller of the first vehicle controls the first vehicle to change lanes to the adjacent lane to avoid the second vehicle, so that the second vehicle can safely drive on the current lane.

[0063] Further, the avoidance condition of the first vehicle not only includes that the available avoidance space needs to meet the avoidance required space of the first vehicle, but also needs to meet that the first vehicle will not collide with other vehicles driving on the adjacent lane when avoiding by lane changing.

[0064] S800, if the available avoidance space does not meet the avoidance condition, control the first vehicle to avoid the second vehicle by tilting.

[0065] Specifically, if the available avoidance space does not meet the avoidance condition, the first vehicle can be controlled to avoid the second vehicle by tilting, that is, the first vehicle does not completely change lanes, but controls the vehicle to tilt at a certain angle on the current lane to leave the maximum space for the second vehicle to pass. In the process of controlling the first vehicle to avoid the second vehicle by tilting, the control is not constrained by the lane line, but needs to avoid collision with vehicles on the adjacent lane.

[0066] In one embodiment, after the avoidance operation of the first vehicle to the second vehicle is completed, the first vehicle can continue to normally drive forward or resume driving on the previous lane, which is not limited here.

[0067] In the vehicle avoidance method, when it is detected that a second vehicle with a higher priority exists in the surrounding range of the first vehicle, whether the first vehicle needs to avoid the second vehicle is accurately determined according to the first driving information of the first vehicle and the second driving information of the second vehicle, and when the first vehicle needs to avoid the second vehicle, available avoidance space in the surrounding range is detected; if the available avoidance space meets the avoidance condition of the first vehicle, the first vehicle is controlled to avoid the second vehicle by changing lanes; if the available avoidance space does not meet the avoidance condition, the first vehicle is controlled to avoid the second vehicle by tilting, and in the whole process, the first driving information of the first vehicle, the second driving information of the second vehicle, and the available avoidance space in the surrounding range are used to accurately determine whether the second vehicle is avoided by changing lanes or by tilting from the driving state of the first vehicle and the second vehicle and the surrounding environment state of the first vehicle, and the maximum passing space is provided for the second vehicle with a higher priority in different situations.

[0068] In an exemplary embodiment, the vehicle avoidance method further comprises:

[0069] The first speed, the first position information, and the first lane information of the first vehicle are extracted from the first driving information, and the second speed, the second position information, and the second lane information of the second vehicle are extracted from the second driving information; the relative position information between the first vehicle and the second vehicle is determined according to the first position information, the first lane information, the second position information, and the second lane information; and whether the first vehicle needs to avoid the second vehicle is determined according to the relative position information, the first speed, and the second speed.

[0070] The driving information at least includes lane information, direction information, speed, and position information of the corresponding vehicle. The relative position information includes relative lane information and relative direction information.

[0071] Specifically, the first speed, the first position information, and the first lane information of the first vehicle are extracted from the first driving information of the first vehicle, the second speed, the second position information, and the second lane information of the second vehicle are extracted from the second driving information of the second vehicle, then the relative direction information between the first vehicle and the second vehicle is determined according to the first position information and the second position information, and the relative lane information between the first vehicle and the second vehicle is determined according to the first lane information and the second lane information. The relative position information between the first vehicle and the second vehicle is obtained by combining the relative direction information and the relative lane information.

[0072] Then, whether the first vehicle needs to avoid the second vehicle is determined according to the relative position information, the first speed, and the second speed.

[0073] In the above embodiments, the relative position information between the first vehicle and the second vehicle, the first speed of the first vehicle, and the second speed of the second vehicle are detected from the first driving information and the second driving information, and then, according to the relative position information, the first speed, and the second speed, it can be accurately determined whether the first vehicle needs to control the second vehicle to give way.

[0074] In an exemplary embodiment, determining whether the first vehicle needs to control the second vehicle to give way according to the relative position information, the first speed, and the second speed comprises:

[0075] If the relative position information indicates that the second vehicle is in the adjacent lane of the lane where the first vehicle is located, the first vehicle does not need to control the second vehicle to give way.

[0076] If the relative position information indicates that the first vehicle and the second vehicle are in the same lane, the second vehicle is in the front area of the driving route of the first vehicle, and the first speed is higher than the second speed, the first vehicle needs to control the second vehicle to change lanes and overtake.

[0077] If the relative position information indicates that the first vehicle and the second vehicle are in the same lane, the first vehicle is in the front area of the driving route of the second vehicle, and the first speed is lower than the second speed, the first vehicle needs to control the second vehicle to give way.

[0078] Specifically, first, it is determined whether the first vehicle needs to control the second vehicle to give way according to the relative position information between the first vehicle and the second vehicle.

[0079] When the relative lane information in the relative position information indicates that the second vehicle is in the adjacent lane of the lane where the first vehicle is located, the first vehicle cannot affect the driving of the second vehicle, and the first vehicle does not need to control the second vehicle to give way. At this time, if the first vehicle needs to change lanes to the lane where the second vehicle is located, it cannot overtake and change lanes to the lane where the second vehicle is located, but needs to adopt a speed reduction lane changing strategy, which focuses on not interfering with the passage of the second vehicle and ensuring the safety of itself and the surrounding environment. That is, the speed of the first vehicle is first reduced to a preset speed reduction threshold, and it is confirmed that there is no collision with the vehicles in the lane where the second vehicle is located and no interference with the passage of the second vehicle before controlling the first vehicle to change lanes to the lane where the second vehicle is located.

[0080] When the relative lane information in the relative position information indicates that the first vehicle and the second vehicle are in the same lane, it cannot be determined at this time whether the first vehicle needs to control the second vehicle to give way, and further, the relative position information, the first speed of the first vehicle, and the second speed of the second vehicle are relied on to determine whether to give way.

[0081] At this time, when it is judged according to the relative position information that the second vehicle is in the front area of the driving route of the first vehicle, if the first speed is higher than the second speed, it means that the first vehicle is about to catch up with the second vehicle, and thus it is necessary to control the first vehicle to change lanes and overtake the second vehicle to avoid the first vehicle meeting the second vehicle in the current lane. Further, changing lanes and overtaking refers to controlling the first vehicle to change lanes to the adjacent lane of the lane where the first lane is located and overtake, at this time, it can be judged whether the second vehicle needs to be overtaken, and the overtaking judgment time can be lengthened to twice the preset overtaking judgment time when the second vehicle is not detected, that is, before overtaking the second vehicle, if there is an overtaking space in the adjacent lane of the lane where the first lane is located, the ego vehicle should provide a certain time for the second vehicle to select a driving trajectory, and only when it is determined that the second vehicle does not change lanes, the ego vehicle changes lanes to the adjacent lane of the lane where the second vehicle is located. Generally, the left lane of the lane where the second vehicle is located is preferred for changing lanes and overtaking.

[0082] When it is judged according to the relative position information that the second vehicle is in the rear area of the driving route of the first vehicle, and the first speed is lower than the second speed, it means that the second vehicle is about to catch up with the first vehicle, and thus it is necessary to control the first vehicle to avoid the second vehicle to make the second vehicle have priority. At this time, the avoidance space required by the first vehicle can be detected to control the first vehicle to change lanes to avoid the second vehicle, or to control the first vehicle to tilt to avoid the second vehicle. Generally, if lane changing is selected, the right lane of the lane where the second vehicle is located is preferred for lane changing.

[0083] In the above embodiment, the relative lane information is first used to judge whether the second vehicle is in the adjacent lane of the lane where the first vehicle is located, and whether the first vehicle needs to be controlled to avoid the second vehicle. If the first vehicle does not need to be controlled to avoid the second vehicle, there is no need to compare the first speed and the second speed, which improves the efficiency of the avoidance judgment. If the relative lane information indicates that the first vehicle and the second vehicle are in the same lane, the relative position information and the comparison between the first speed and the second speed are used to accurately judge whether the first vehicle needs to be controlled to change lanes and overtake the second vehicle or tilt to avoid the second vehicle.

[0084] In an exemplary embodiment, the first driving information includes a first speed of the first vehicle; as shown in Figure 3 S800 includes:

[0085] S820, when the available avoidance space does not meet the avoidance condition, and the first speed is less than or equal to a preset driving speed threshold, the minimum allowed distance between the first vehicle and other vehicles is obtained.

[0086] S840, based on the minimum allowed distance matched tilt angle, the first vehicle is controlled to tilt to the adjacent lane.

[0087] The minimum allowable distance between the first vehicle and other vehicles means that the distance between the first vehicle and other vehicles within the vehicle's perimeter cannot be less than the minimum allowable distance.

[0088] Specifically, when the first speed is less than or equal to a preset driving speed threshold, the first vehicle is judged to be in a crowded vehicle environment and is moving slowly. Even if the first vehicle continues to move forward, it is difficult to detect available avoidance space that meets the first vehicle's avoidance conditions. At this time, the first vehicle is in the area ahead of the second vehicle's driving path. The minimum allowable distance between the first vehicle and the other vehicle is then obtained. Once the distance between the first vehicle and a vehicle traveling in an adjacent lane is less than the minimum allowable distance, the first vehicle may be at risk of collision with the vehicle traveling in the adjacent lane. Therefore, a tilt angle that matches the minimum allowable distance can be determined. Based on the tilt angle that matches the minimum allowable distance, the first vehicle is controlled to tilt toward the adjacent lane. For example, the first vehicle is controlled to tilt to the left or right by a preset angle to avoid collision with vehicles traveling in the adjacent lane when the first vehicle tilts by the preset angle. In actual applications, the preset tilt angle is determined based on the minimum allowable distance and the position information of the first vehicle. Generally, the tilt angle defaults to 45 degrees.

[0089] When controlling the first vehicle to move obliquely toward the adjacent lane, the first vehicle travels without being affected by the constraints of the lane lines to provide the second vehicle with the maximum passing space. At the same time, the vehicle controller of the first vehicle will also push an oblique avoidance reminder message to other vehicles within a preset space range that have the same passing priority as the first vehicle or a lower passing priority than the first vehicle. The oblique avoidance reminder message is used to call on other vehicles within the vehicle surrounding range of the first vehicle to perform vehicle avoidance. The oblique avoidance reminder message includes but is not limited to auditory reminder message and visual reminder message. For example, when the oblique avoidance reminder message is an auditory reminder message, the first vehicle may use the on-board microphone to make a voice broadcast reminder to other vehicles to avoid the vehicle. For another example, when the oblique avoidance reminder message is a visual reminder message, the first vehicle may flash the on-board work lights to remind other vehicles that they need to avoid the vehicle itself and the second vehicle.

[0090] In the above embodiment, when the driving speed is less than or equal to the preset driving speed threshold, that is, when the first vehicle is in a congested area, it is impossible to speed up to re-detect the available avoidance space within the vehicle's perimeter. Therefore, the avoidance method can be changed to control the first vehicle to tilt and avoid the second vehicle, so that the first vehicle can provide the second vehicle with the maximum passing space even when it is unable to change lanes.

[0091] In an exemplary embodiment, the first driving information includes a first speed of the first vehicle; and the vehicle avoidance method further includes:

[0092] If the available avoidance space does not satisfy the avoidance condition and the first speed is greater than the preset driving speed threshold, the driving speed of the first vehicle is increased, and the step of detecting the available avoidance space in the surrounding range of the vehicle is returned until the detected available avoidance space satisfies the avoidance condition.

[0093] Specifically, when the available avoidance space does not satisfy the avoidance condition, the driving speed of the first vehicle can be extracted from the first driving information, and then based on the first speed of the first vehicle, it is judged how to determine the avoidance manner of the first vehicle to the second vehicle in the case that the second vehicle cannot be lane-changing avoided at present.

[0094] It should be noted that based on the driving speed of the first vehicle, it can be judged whether the surrounding environment state of the first vehicle is in a vehicle crowded environment, and therefore the determined avoidance manner is actually an avoidance manner matched with the surrounding environment state.

[0095] In the above embodiment, it is pointed out that when the first speed is less than or equal to the preset driving speed threshold, the first vehicle is controlled to perform inclined avoidance to the second vehicle, and when the first speed is greater than the preset driving speed threshold, it indicates that the first vehicle is not in a vehicle crowded environment at this time, and since the current driving speed cannot detect the available avoidance space satisfying the avoidance condition, the driving speed of the first vehicle can be increased to control the first vehicle to continue driving and wait for the available avoidance space satisfying the avoidance condition to be detected to avoid the second vehicle, that is, after the driving speed of the first vehicle is increased, the step of detecting the available avoidance space in the surrounding range of the vehicle is returned until the detected available avoidance space satisfies the avoidance condition, and then the first vehicle is controlled to perform lane-changing avoidance to the second vehicle. In practical application, the lane where the first vehicle is located is provided with a maximum driving speed threshold, and when the driving speed of the first vehicle is increased, the driving speed cannot exceed the maximum driving speed threshold.

[0096] In the above embodiment, if the available avoidance space does not satisfy the avoidance condition and the first speed is greater than the preset driving speed threshold, the driving speed of the first vehicle can be increased, and then whether the available avoidance space in the surrounding range of the vehicle satisfies the avoidance condition of the first vehicle is detected in real time until the detected available avoidance space satisfies the avoidance condition, and then the first vehicle is controlled to perform lane-changing avoidance to the second vehicle, that is, if the available avoidance space does not satisfy the avoidance condition and the first speed is greater than the preset driving speed threshold, the first vehicle needs to continue to speed up and constantly detect the available avoidance space, and wait for the detected available avoidance space to satisfy the avoidance condition.

[0097] In one exemplary embodiment, as shown in Figure 4 the vehicle avoidance method further comprises:

[0098] S862, if the available avoidance space does not satisfy the avoidance condition and the first speed is greater than the preset driving speed threshold, start timing whether the available avoidance space satisfying the avoidance condition is detected.

[0099] S864, if the available avoidance space satisfying the avoidance condition is not detected within the first preset time range, trigger an avoidance message notification control of the first vehicle, wherein the avoidance message notification control is configured to provide avoidance reminding information and push the avoidance reminding information to other vehicles.

[0100] S866, if the available avoidance space satisfying the avoidance condition is detected within the second preset time range, control the first vehicle to perform lane change avoidance on the second vehicle, wherein the second preset time range is longer than the first preset time range.

[0101] S868, if the available avoidance space satisfying the avoidance condition is not detected within the second preset time range, push a manual driving prompt information to a driver terminal of the first vehicle.

[0102] Specifically, if the available avoidance space does not satisfy the avoidance condition and the first speed is greater than the preset driving speed threshold, at this time the first vehicle needs to continue to speed up and constantly detect the available avoidance space in the vehicle range to wait for the detected available avoidance space to satisfy the avoidance condition. Therefore, at this time, the timing whether the available avoidance space satisfying the avoidance condition is detected can be started.

[0103] In the process of continuing to drive the first vehicle, if the available avoidance space satisfying the avoidance condition is not detected within the first preset time range, the vehicle controller of the first vehicle will trigger the avoidance message notification control of the first vehicle, that is, generate avoidance reminding information and push the avoidance reminding information to the vehicles within the vehicle range of the first vehicle within the preset space range, so that the vehicles within the vehicle range of the first vehicle know that the first vehicle will avoid the second vehicle and need to provide certain avoidance space for the first vehicle. In practical applications, the avoidance reminding information includes but is not limited to auditory reminding information such as horn and voice broadcast, and visual reminding information such as working lamp flickering in a specific way.

[0104] However, sometimes even if the avoidance reminding information is pushed, after a certain time, for example, the available avoidance space satisfying the avoidance condition is still not detected within the second preset time range, at this time the first vehicle still cannot perform lane change avoidance, and a manual driving prompt information needs to be pushed to the driver terminal of the first vehicle to prompt the driver of the first vehicle to take over manually.

[0105] If the available avoidance space satisfying the avoidance condition is detected within the second preset time range, the first vehicle is controlled to perform lane change avoidance on the second vehicle.

[0106] For example, when the driving speed of the first vehicle is greater than or equal to 15 Kph, the first vehicle is preferred to adopt a speed-up lane-changing avoidance scheme based on the surrounding environment, and attempts to drive into the adjacent lane of the lane where the first vehicle is located; when no available avoidance space meeting the avoidance condition is detected for 10 seconds, the first vehicle will automatically trigger the horn and the voice outside the vehicle to inform other vehicles to avoid; if no available avoidance space meeting the avoidance condition is detected for a long time (greater than or equal to 30 seconds), the driver of the first vehicle will be prompted to take over manually.

[0107] In the above embodiment, when the first speed of the first vehicle is greater than the preset driving speed threshold, the first vehicle continues to speed up and drive, and if no available avoidance space meeting the avoidance condition is detected within the first preset time range after starting to speed up and drive, other vehicles can be reminded to reserve avoidance space for the first vehicle, and if no available avoidance space meeting the avoidance condition is detected until the second preset time range, the driver needs to be prompted to take over manually. By setting two time ranges, the efficiency of controlling the first vehicle to avoid the second vehicle is improved.

[0108] In an exemplary embodiment, before obtaining the first driving information of the first vehicle and the second driving information of the second vehicle, the vehicle avoidance method further comprises:

[0109] The vehicle type recognition result of each to-be-recognized vehicle in the vehicle surrounding range of the first vehicle is detected by at least one of a visual recognition method and an auditory recognition method; and whether the second vehicle exists in the vehicle surrounding range of the first vehicle is detected according to the vehicle type recognition result of each to-be-recognized vehicle.

[0110] Specifically, whether the second vehicle exists in the vehicle surrounding range of the first vehicle is detected by the visual recognition method, the auditory recognition method, and the fusion of the visual recognition method and the auditory recognition method. That is, the vehicle type recognition result of each to-be-recognized vehicle in the vehicle surrounding range of the first vehicle is detected by the visual recognition method, the auditory recognition method, and the fusion of the visual recognition method and the auditory recognition method, and then whether the second vehicle exists in the vehicle surrounding range of the first vehicle is detected according to the vehicle type recognition result of each to-be-recognized vehicle. In some embodiments, detecting the vehicle type recognition result of each to-be-recognized vehicle in the vehicle surrounding range of the first vehicle does not mean that the specific type of each to-be-recognized vehicle will be detected, but whether each to-be-recognized vehicle is a vehicle type with a higher priority than the first vehicle.

[0111] In practical applications, the visual recognition method includes but is not limited to feature recognition such as vehicle structure and vehicle color, and other vehicle motion trend recognition of the vehicle surrounding range, and the auditory recognition method includes but is not limited to vehicle audio spectrum recognition and vehicle voice broadcast.

[0112] In one embodiment, when the second vehicle is located by integrating visual recognition and auditory recognition, a primary recognition scheme and an auxiliary recognition scheme can be determined. Since the visual recognition method is more accurate than the auditory recognition method, generally, the visual recognition method is determined as the primary recognition scheme, and the auditory recognition method is determined as the auxiliary recognition scheme.

[0113] At this time, when one of the visual recognition method or the auditory recognition method fails to capture the vehicle type recognition result of the vehicle to be identified, the result of the other recognition method will be output as the vehicle type recognition result of the vehicle to be identified; when the visual recognition method or the auditory recognition method both effectively identify the vehicle type recognition result of the vehicle to be identified, but the vehicle type recognition results of the vehicle to be identified output by the two recognition methods are inconsistent, the recognition result of the visual recognition method shall prevail.

[0114] In the above embodiment, the vehicle type identification result of each vehicle to be identified within the vehicle surrounding range of the first vehicle is detected by at least one of visual recognition and auditory recognition. Compared with the single recognition method in the prior art, this is more accurate and improves the accuracy of detecting whether there is a second vehicle within the vehicle surrounding range of the first vehicle.

[0115] In an exemplary embodiment, detecting a vehicle type recognition result of each to-be-identified vehicle within a perimeter of the first vehicle by visual recognition includes:

[0116] Acquire a video frame image within the vehicle perimeter of the first vehicle; extract vehicle visual feature information of each vehicle to be identified within the vehicle perimeter of the first vehicle from the video frame image, and detect a vehicle type recognition result of each vehicle to be identified within the vehicle perimeter of the first vehicle based on the vehicle visual feature information.

[0117] Specifically, each first vehicle is equipped with a visual detection device, which may be a camera, a photographic device, an infrared sensor, or the like. For example, if the visual detection device is a camera, multiple cameras are evenly installed around the first vehicle to capture multiple video frames within the perimeter of the vehicle and perform feature extraction on each captured video frame.

[0118] That is, at least one vehicle visual feature information of each to-be-identified vehicle in the vehicle surrounding range of the first vehicle is extracted from all video frame images, and the at least one vehicle visual feature information is compared with corresponding feature information of a vehicle with a higher priority, to detect a vehicle type identification result of each to-be-identified vehicle in the vehicle surrounding range of the first vehicle, and to determine whether each to-be-identified vehicle belongs to a vehicle type with a higher priority than the priority of the ego vehicle. For example, when the vehicle type identification result of the to-be-identified vehicle is a police vehicle, it is considered that the priority of the to-be-identified vehicle is higher than the priority of the ego vehicle, and it is considered that the second vehicle is detected.

[0119] In one embodiment, in the process of extracting vehicle visual feature information of each to-be-identified vehicle in the vehicle surrounding range of the first vehicle from the video frame images, first, each to-be-identified vehicle in the vehicle surrounding range of the first vehicle needs to be determined from the video frame images, and then the vehicle visual feature information of each to-be-identified vehicle is extracted.

[0120] In the process of determining each to-be-identified vehicle in the vehicle surrounding range of the first vehicle from each video frame image, the drivable area passing road shoulder in the video frame image can be distinguished as a target, where the road shoulder refers to a strip structure part with a certain width located from the edge of the carriageway to the edge of the roadbed. Generally, the drivable area passing road shoulder in the video frame image can be distinguished as a target through a deep learning target detection framework, such as a Region-CNN (Region-Based Convolutional Neural Network) target detection framework, a Faster R-CNN (Faster Region-Based Convolutional Neural Network) target detection framework, etc. Then, the to-be-identified vehicle is extracted as a POI (Point of Interest) in the obtained drivable area by using a pre-trained Region-CNN, Faster R-CNN, etc. operator, so as to accurately identify the to-be-identified vehicle from the video frame image.

[0121] In an embodiment, before extracting the vehicle visual feature information of each to-be-identified vehicle in the vehicle surrounding range of the first vehicle from the video frame image, each video frame image needs to be preprocessed, and the preprocessing process includes but is not limited to noise reduction preprocessing, contrast enhancement preprocessing, and size optimization adjustment preprocessing. In actual application, the noise reduction preprocessing includes but is not limited to median filtering, Gaussian filtering and the like, the contrast enhancement preprocessing includes but is not limited to color histogram equalization, linear contrast stretching and the like, and the size optimization adjustment preprocessing includes but is not limited to interpolation scaling algorithm, scaling processing on the interest area through a sliding window and the like.

[0122] In an embodiment, the vehicle visual feature information includes but is not limited to vehicle structure feature information and vehicle color feature information, and the vehicle color feature information includes but is not limited to body color feature information and alarm lamp color feature information. The vehicle structure feature information includes but is not limited to body text feature information, alarm lamp structure feature information and body logo feature information. The vehicle type identification result of each to-be-identified vehicle in the vehicle surrounding range of the first vehicle can be detected through any one of the vehicle visual feature information.

[0123] For the body color feature information, each video frame image is subjected to RGB (Red, Green, Blue) body distribution extraction, and then is converted to Lab (Lightness Green-Red Axis-Blue Yellow Axis) space through RGB component. The body color occupying more than 80% of the area in the vehicle contour area is obtained as the body main color, and the coordinates [L1, a1, b1] of the body main color in the Lab space are obtained. By comparing the measured [L1, a1, b1] with the preset body main color [L 1标准 , a 1标准 , b 1标准 ] of the vehicle type with higher traffic priority, if the measured [L 1标准 , a 1标准 , b 1标准 ] is within [L 1标准 , a 1标准 , b 1标准 ], the body color domain verification is valid, and the vehicle is output as the matched vehicle type with higher traffic priority. In actual application, [L 1标准 , a 1标准 , b 1标准 ] is a range value. Taking police cars, ambulances, fire engines and military vehicles as examples, the range information of [L 1标准 , a 1标准 , b 1标准 ] of various vehicle types is shown in Table 1.

[0124]

[0125] For the alarm light color feature information, the RGB body distribution of each video frame image is extracted respectively, and then converted to Lab space through RGB component. The rectangular protruding structure in the roof part of the vehicle contour area is defined as the top alarm light. Generally, the width of the rectangular protruding structure is required to be more than 50% of the total width of the vehicle body, and the coordinates [L2, a2, b2] of the top alarm light color in the Lab space are obtained. By comparing with the preset top alarm light color [L 2标准 , a 2标准 , b 2标准 ] of the vehicle type with higher traffic priority respectively, if the measured [L 2标准 , a 2标准 , b 2标准 ] is within [L 2标准 , a 2标准 , b 2标准 ], the top alarm light color domain verification is valid, and the vehicle is output as the matched vehicle type with higher traffic priority. In practical application, [L 2标准 , a 2标准 , b 2标准 ] is a range value. Taking police cars, ambulances, fire engines and military vehicles as examples, the range information of [L 2标准 , a 2标准 , b 2标准 ] of various vehicle types is shown in Table 2 as follows:

[0126]

[0127] When it is monitored that the color domain component of the top alarm light of the vehicle to be identified is within the valid interval of [L 1标准 , a 1标准 , b 1标准 ], it is judged that the top alarm light is in working state, and the working result of the alarm light is output.

[0128] For the body text feature information, character recognition is performed on each video frame image in the vehicle body area. The character recognition means include but are not limited to traditional OCR (Optical Character Recognition) text detection scheme, end-to-end OCR model, etc. After completing the character recognition, if the vehicle body text belongs to the special vehicle identification library representing the vehicle type identification library with higher traffic priority, it is determined that the vehicle to be identified is the vehicle type with higher traffic priority. For example, if the vehicle body text belongs to the special vehicle identification library, the vehicle to be identified is output as a special vehicle. Generally, the special vehicle identification library includes but is not limited to “court”, “police”, “ambulance”, “command vehicle” and other special vehicle name identification.

[0129] For the alarm light structure feature information, since the vehicle types with high traffic priority are limited to the specifications of the alarm light, different views of the alarm light of the identified vehicle can be obtained in different orientations. At this time, edge detection is performed on each video frame image, and the edge detection methods include but are not limited to Canny operator-based edge detection, Prewitt operator-based edge detection, etc. The detected edge features are extracted, and the target contour in the video frame image is identified as the front, back, left and right directions of the vehicle. When the vehicle to be identified faces the first vehicle in the left and right directions, it is determined whether the vehicle to be identified belongs to a special vehicle by judging whether the top of the vehicle to be identified protrudes a circular feature (i.e., the X and Y sizes are less than a certain proportion, and there is no corner feature). If the top of the vehicle to be identified protrudes a circular feature, the vehicle to be identified belongs to a special vehicle. When the vehicle to be identified faces the first vehicle in the front and back directions, it is determined whether the vehicle to be identified belongs to a special vehicle by judging whether the top of the vehicle to be identified has a rectangular feature (i.e., the X and Y sizes are less than a certain proportion, and there is a corner feature). When the top of the vehicle to be identified has a rectangular feature, it is determined that the vehicle to be identified belongs to a special vehicle.

[0130] For the vehicle body and logo feature information, for each video frame image, the tail of each vehicle to be identified in the video frame image is located in the license plate area according to color (such as blue, yellow, green, black, etc.) and rectangular shape. In the area between the top of the license plate area and the upper boundary of the tailgate, the vehicle logo feature recognition is completed by the sparse sampling gray scale distribution and gradient distribution feature method. The detailed process of the sparse sampling gray scale distribution and gradient distribution feature method is as follows: in the area between the top of the license plate area and the upper boundary of the tailgate, random uniform sparse sampling is performed, the sampling points are divided into adjacent points and non-adjacent points, the gradient amplitudes of 8 angle ranges of the adjacent points are extracted as gradient distribution features, and the brightness relationship of the non-adjacent points is calculated as gray scale distribution features. Finally, the above two types of features are extracted by scaling the window scanning the detection area, and are matched with the information in the feature library in reverse to determine the logo type of the vehicle to be identified, and then the vehicle visual feature information of the vehicle to be identified is detected according to the logo type of the vehicle to be identified.

[0131] In the above embodiment, at least one of the vehicle body color feature information, the alarm light color feature information, the vehicle body text feature information, the alarm light structure feature information and the vehicle body and logo feature information of each vehicle to be identified in the surrounding range of the first vehicle is extracted from the video frame image, and the vehicle type identification result of each vehicle to be identified in the surrounding range of the first vehicle is accurately detected to determine whether the first vehicle needs to avoid other vehicles.

[0132] In an exemplary embodiment, the vehicle type identification result of each to-be-identified vehicle in the surrounding range of the first vehicle is detected by an auditory identification manner, including:

[0133] When it is detected that the volume of the environmental audio exceeds a preset volume threshold, the environmental sound in the surrounding range of the first vehicle is collected, and the vehicle type identification result of each to-be-identified vehicle in the surrounding range of the first vehicle is detected according to the frequency spectrum of the environmental sound; and / or, the voice prompt information emitted by the remaining vehicles except the first vehicle is monitored, and the vehicle type identification result of each to-be-identified vehicle in the surrounding range of the first vehicle is detected according to the voice prompt information.

[0134] Specifically, the vehicle type identification result of each to-be-identified vehicle in the surrounding range of the first vehicle is detected by an auditory identification manner, wherein the auditory identification manner includes but is not limited to the following two kinds:

[0135] Firstly, the first vehicle is provided with an audio monitoring device and an environmental sound collection device, and the vehicle controller of the first vehicle controls the audio monitoring device to continuously monitor the sound outside the vehicle. Since the vehicles with higher passing priority generally have sound pressure requirements, the auditory effectiveness triggering judgment can be performed by using a fixed volume threshold, that is, when it is detected that the volume of the environmental audio exceeds the preset volume threshold, the vehicle controller of the first vehicle starts to control the environmental sound collection device to collect the environmental sound; when it is detected that the volume of the environmental audio does not exceed the preset volume threshold, the first vehicle temporarily hibernates the environmental sound collection device or uses it for other applications, so as to save energy consumption and computing resources.

[0136] When the environmental sound collection device is started, the environmental sound collection device will continuously collect the environmental sound according to a certain time period, store the collected environmental sound in the vehicle-end storage, and send it to the vehicle controller of the first vehicle. The vehicle controller extracts the frequency spectrum of the collected environmental sound, and compares the frequency spectrum extraction result with the standard sound frequency spectrum range of the vehicles with higher passing priority which is stored in advance. If the frequency spectrum extraction result is within the standard sound frequency spectrum range, it is judged that the collected environmental sound is effective. At this time, the vehicle type identification result of each to-be-identified vehicle in the surrounding range of the first vehicle is detected according to the collected environmental sound, so as to position the second vehicle.

[0137] In practical applications, the spectrum extraction result includes but is not limited to power spectral density, frequency distribution, audio decibel, audio period, etc. The means for performing spectrum extraction on the collected environmental sound can be a method of using mel frequency cepstral coefficient or a method of Fourier transform, etc. According to the collected environmental sound, the method for detecting the vehicle type recognition result of each to-be-identified vehicle in the vehicle surrounding range of the first vehicle includes: using the body-symmetrical arranged external MIC (Microphone, microphone), the body-symmetrical arranged external MIC can select symmetrical longitudinal arrangement of even number of MICs, through the received environmental sound and the corresponding MIC arrangement position, the system performs boost position calculation according to the sound pressure direction of the environmental sound in the first vehicle rear, left rear, right rear, front, left front and right front, to determine the positioning result of the second vehicle, and according to the positioning result of the second vehicle, the vehicle type recognition result of each to-be-identified vehicle in the vehicle surrounding range of the first vehicle is detected. At this time, the vehicle type recognition result can be a vehicle type with higher priority for passing or a common vehicle type.

[0138] Secondly, the avoidance voice prompt information emitted by the remaining vehicles except the first vehicle is monitored. For the remaining vehicles in the vehicle surrounding range and close to the first vehicle, the voice prompt information emitted by the remaining vehicles is monitored. If the voice prompt information is recognized to indicate that the ego vehicle avoids a certain vehicle, it means that there is a vehicle type with higher priority for passing in the vehicle surrounding range of the first vehicle. At this time, the voice prompt information also contains at least one of the position information and the license plate information of the second vehicle. Therefore, according to at least one of the position information and the license plate information of the second vehicle appearing in the voice prompt information, the vehicle controller of the first vehicle determines the positioning result of the second vehicle, and according to the positioning result of the second vehicle, the vehicle type recognition result of each to-be-identified vehicle in the vehicle surrounding range of the first vehicle is detected. In practical applications, the voice prompt information indicating that the ego vehicle avoids a certain vehicle refers to the avoidance semantic information such as "avoidance" and "pull over" in the voice prompt information.

[0139] In one embodiment, the extraction step of mel frequency cepstral coefficient can be divided into 9 key steps, which are designed based on the human ear hearing model to ensure that the features can reflect the essential characteristics of the speech. The following is the standard process:

[0140] S1, pre-emphasis:

[0141] Pre-emphasis Analog human ear is sensitive to high frequencies; its purpose is to enhance high-frequency components (high-frequency energy in speech signals is usually weak), avoid numerical problems in FFT (Fast Fourier Transform) calculation, and improve signal-to-noise ratio; the method includes: using a first-order high-pass filter, the formula is: ( y(t) = x(t) - \alpha x(t-1) ), where ( \alpha ) is usually 0.95 or 0.97, which can effectively enhance the intelligibility of speech.

[0142] S2, Framing:

[0143] The frame length and frame shift of framing need to be adjusted according to the application; its purpose is that speech signals are non-stationary, and framing cuts the signal into short-time segments (frames), making each frame approximately stationary, which is convenient for subsequent processing. The method includes: the frame length is usually 25ms (for example, at a sampling rate of 16kHz, one frame has 400 samples), and the frame shift is 10ms (overlapping 50%), which ensures the continuity between frames.

[0144] S3, Windowing:

[0145] Windowing can optimize spectral estimation and improve feature stability; its purpose is to reduce spectral leakage caused by framing and smooth the frame edges; the method includes: applying a window function (such as Hamming window or Hanning window), the formula is: ( w(n) = 0.54 -0.46 \cos\left(\frac{2\pin}{N-1}\right) ), where N is the frame length, and "\ " is to let the subsequent symbols follow the mathematical formula specification format.

[0146] S4, Fast Fourier Transform:

[0147] Fast Fourier Transform is the premise of power spectrum calculation; its purpose is to convert time-domain signals into frequency-domain and obtain spectral information. The method includes: performing Fast Fourier Transform on each frame of signal to obtain complex spectrum, usually the number of Fast Fourier Transform points is set to 1024 or 512 to cover the audible range of human ear.

[0148] S5, Power spectrum estimation:

[0149] Power spectrum is the input of Mel (Mel-frequency Scale) filter bank; its purpose is to calculate the energy distribution of the signal, which is used for subsequent filtering. The method includes: taking the absolute value square of the FFT result to obtain the power spectrum ( |X(k)|^2 ), where ( k ) is the frequency index.

[0150] S6, Mel filter bank application:

[0151] Mel filter bank is the core of MFCC, which maps linear spectrum to perceptually relevant Mel domain; the purpose is to simulate the nonlinear frequency perception of human ear (Mel scale), higher resolution in low frequency part and lower in high frequency. The method includes: design a set of triangular filters (usually 20-40), covering 0Hz to Nyquist frequency (e.g. 8kHz), filters are uniformly distributed in Mel frequency: Mel frequency formula is ( \text{Mel}(f) = 2595 \log_{10}(1 + f / 700) ), multiply power spectrum with each filter, sum up to get the energy of each filter ( E_m ).

[0152] S7, take logarithm:

[0153] Logarithmic operation simulates the compression characteristics of hearing; the purpose is to compress the dynamic range (human ear's perception of sound intensity is approximately logarithmic), enhance feature robustness; the method includes: taking natural logarithm of the energy of each filter: ( \log(E_m) ), get log Mel spectrum.

[0154] S8, discrete cosine transform:

[0155] Discrete cosine transform produces cepstral coefficients; the purpose is to convert log Mel spectrum to cepstral domain, decorrelate features, reduce dimensionality; the method includes: apply discrete cosine transform, formula is: ( c_n = \sum_{m=0}^{M-1} \log(E_m) \cos\left(\frac{\pi n(m + 0.5)}{M}\right) ), where (M) is the number of filters, (n) is the coefficient index, usually keep the first 2 to 13 coefficients (12 coefficients), discard the first (representing the direct current component) and higher order coefficients.

[0156] S9: feature normalization (optional):

[0157] Normalization is an optimization step, especially in noisy environments; the purpose is to improve noise robustness and model performance; the method includes: mean normalization or variance scaling of Mel frequency cepstral coefficients, commonly used in practical applications.

[0158] In the above embodiment, by various hearing recognition methods, the vehicle type recognition result of each to-be-identified vehicle in the vehicle surrounding range of the first vehicle is detected, which can assist in detecting the vehicle type recognition result of each to-be-identified vehicle in the vehicle surrounding range of the first vehicle while visually recognizing.

[0159] In one exemplary embodiment, the vehicle type recognition result of each to-be-identified vehicle in the vehicle surrounding range of the first vehicle is detected by the visual recognition method and the hearing recognition method, including:

[0160] When the second vehicle is not detected through the vehicle visual feature information, the area information of the second vehicle is predicted through the hearing recognition mode; the first lane number of the vehicles changing lanes from the current lane of the first vehicle to the adjacent lane of the current lane in the target direction of the first vehicle is monitored, and the second lane number of the vehicles changing lanes from the adjacent lane to the current lane is monitored; the lane information of the second vehicle is detected according to the first lane number and the second lane number; and the vehicle type recognition result of each vehicle to be recognized in the vehicle surrounding range of the first vehicle is obtained according to the lane information and the area information.

[0161] Specifically, when the second vehicle is detected through the vehicle visual feature information, the second vehicle can be detected according to the visual recognition mode in the above embodiment, and the second driving information of the second vehicle is output.

[0162] When the second vehicle is not detected through the vehicle visual feature information, it is considered that the second vehicle may be blocked by other vehicles, so the area information of the second vehicle can be predicted through the hearing recognition mode first. The hearing recognition mode is described above and will not be repeated here.

[0163] Since the positioning result obtained by the hearing recognition mode is not accurate enough, a rough area prediction information is often obtained, so when the vehicle changing lanes from the current lane to the adjacent lane in the target direction of the first vehicle is monitored, the first lane number of the vehicles changing lanes from the current lane of the first vehicle to the adjacent lane of the current lane in the target direction of the first vehicle is monitored through the shooting device within a preset time period, and the second lane number of the vehicles changing lanes from the adjacent lane to the current lane within the preset time period is monitored, so as to determine the vehicle movement trend in the target direction of the first vehicle according to the first lane number and the second lane number. When the ratio between the first lane number and the second lane number is higher than a preset ratio threshold, that is, more vehicles change lanes from the current lane to the adjacent lane, it is considered that there is a second vehicle with higher priority in the target direction of the first vehicle in the current lane. When the ratio between the first lane number and the second lane number is less than the preset ratio threshold, that is, more vehicles change lanes from the adjacent lane to the current lane, it is considered that there is a second vehicle with higher priority in the target direction of the first vehicle in the adjacent lane of the current lane. Since the second vehicle in the adjacent lane does not need to be avoided, the timing of monitoring the lane changing number can be reset. In actual application, the target direction can be the front of the first vehicle on the driving route, or the rear of the first vehicle on the driving route.

[0164] The specific positioning result of the second vehicle is obtained by combining the lane information and the area information, and the vehicle type recognition result of each vehicle to be recognized in the vehicle surrounding range of the first vehicle is obtained.

[0165] In one embodiment, not only in the case where the second vehicle is not detected through the vehicle visual feature information, but also in the case where the vehicle type identification result of each to-be-identified vehicle in the vehicle surrounding range of the first vehicle is detected through the combination of visual identification and auditory identification, for example, when the special vehicle or the special vehicle team is behind the vehicle lane of the ego vehicle, the special vehicle or the special vehicle team is monitored through the visual identification mode in the activated state of the alarm light or the warning light double flashing non-activated state, and the auditory identification mode determines that there is avoidance information in the voice prompt information of the to-be-identified vehicle, the ego vehicle performs avoidance, in addition, the auditory identification mode assists the visual identification mode to monitor the orientation information and the lane information of the second vehicle.

[0166] In the above embodiment, when the second vehicle is not detected through the vehicle visual feature information, the detection mode of the vehicle motion trend in the visual identification mode is combined with the auditory identification mode, and the vehicle type identification result of each to-be-identified vehicle in the vehicle surrounding range of the first vehicle can also be accurately detected.

[0167] In one exemplary embodiment, before the first driving information of the first vehicle and the second driving information of the second vehicle are obtained in the case where the second vehicle exists in the vehicle surrounding range of the first vehicle, the following steps are included:

[0168] The second driving information of the second vehicle sent by the cloud is obtained, wherein the cloud is configured to send the second driving information detected by the vehicle controller of the third vehicle to the vehicle controllers of a plurality of vehicles in a preset spatial range; and the second vehicle in the vehicle surrounding range of the first vehicle is detected according to the second driving information.

[0169] Specifically, each vehicle is configured with a vehicle controller of the ego vehicle, each vehicle controller is in communication connection with the cloud, and each vehicle controller can obtain the second driving information of the second vehicle when the second vehicle is detected, and upload the second driving information of the second vehicle to the cloud. The cloud realizes information sharing of the received second driving information of the second vehicle, that is, the cloud sends the second driving information of the second vehicle to the vehicle controllers of a plurality of vehicles in a preset spatial range, including the vehicle controller of the first vehicle, so that the vehicle controller of the first vehicle can obtain the second driving information of the second vehicle detected by the vehicle controller of the third vehicle, and compare the second driving information of the second vehicle with the driving information of the ego vehicle to detect whether the second vehicle exists in the vehicle surrounding range of the first vehicle.

[0170] Further, the cloud sends the second driving information of the second vehicle to vehicle controllers of a plurality of vehicles within a preset spatial range, not necessarily to vehicle controllers of all vehicles within the preset spatial range, but needs to obtain driving information of all vehicles within the preset spatial range, resolve the driving information of all vehicles within the preset spatial range with the second driving information, determine whether there is a vehicle meeting the second vehicle, and if so, send the second driving information to the corresponding vehicle.

[0171] Taking the second vehicle as a special vehicle as an example, within a certain spatial range, when one of the vehicles in the spatial range detects that a special vehicle appears around the vehicle, the vehicle estimates the positioning information of the vehicle and the relative position of the special vehicle relative to the vehicle, obtains the driving information of the special vehicle such as the positioning position, running direction and movement speed of the special vehicle, uploads the converted driving information of the special vehicle to the cloud storage, and the cloud resolves the meeting situation with the special vehicle based on the running speed and distance of the remaining vehicles within the spatial range, and sends the driving information of the special vehicle and the meeting time to the vehicles that will meet the special vehicle, so that the remaining vehicles can also obtain the detection result of the special vehicle when they cannot effectively monitor the special vehicle within the spatial range.

[0172] In the above embodiment, through the information sharing mechanism of the cloud, the remaining vehicles can efficiently determine whether the second vehicle exists within the vehicle surrounding range of the first vehicle within the preset spatial range even if they cannot effectively monitor the second vehicle.

[0173] In an exemplary embodiment, the vehicle avoidance method further comprises:

[0174] In the process of controlling the first vehicle to avoid the second vehicle by lane changing or tilting, the line of sight direction of the driver of the first vehicle is detected; when the line of sight direction represents that the driver looks ahead, at least one of the first visual prompt information matched with the line of sight direction and the auditory prompt information is provided to the first vehicle; and when the line of sight direction represents that the driver does not look ahead, at least one of the first visual prompt information, the auditory prompt information, the tactile prompt information and the second visual prompt information matched with the line of sight direction is provided to the first vehicle.

[0175] Specifically, all vehicles in the present application are in the intelligent driving control state, and for the intelligent driving control vehicle, the driver should be provided with clear state changes and motion intentions, therefore, the driver needs to be effectively prompted.

[0176] In order to enable the driver to receive effective prompt information, the line-of-sight direction of the driver of the first vehicle can be detected in advance, and it is determined whether the line-of-sight direction of the driver of the first vehicle is looking at the front. For example, when the line-of-sight direction of the driver of the first vehicle is looking up at the steering wheel or the front glass of the vehicle, it is determined that the line-of-sight direction of the driver of the first vehicle is looking at the front.

[0177] When the line-of-sight direction represents that the driver is looking at the front, the vehicle controller of the first vehicle provides the driver with forward prompt information, wherein the forward prompt information includes at least one of auditory prompt information and first visual prompt information matched with the line-of-sight direction. The first visual prompt information includes but is not limited to being displayed on a vehicle front status instrument screen, a head-up display, or a central control large screen. The vehicle front status instrument screen displays a status prompt through text or warning signs, such as: “the vehicle is avoiding a special vehicle”. The head-up display or the central control screen displays an enhanced virtual area of the intelligent driving vehicle environment. At the starting time point of the system performing lane changing of the special vehicle, the virtual area of the intelligent driving vehicle environment is zoomed out from the vehicle to cover a wider road area to display the second vehicle approaching the surrounding vehicle, and the second vehicle is rendered according to the recognized vehicle color, warning light, and warning light working state. Meanwhile, when the second vehicle is monitored within the surrounding range of the first vehicle, the cabin plays a voice prompt that the special vehicle is approaching in the direction of the approaching second vehicle through auditory prompt information. When the second vehicle triggers lane changing, the system also provides a voice prompt: “xx vehicle is performing lane changing of the second vehicle”.

[0178] When the line-of-sight direction represents that the driver is not looking at the front, not only the above-mentioned forward prompt information needs to be provided, but also at least one of first vehicle tactile prompt information and second visual prompt information matched with the line-of-sight direction needs to be added. The second visual prompt information matched with the line-of-sight direction includes but is not limited to lateral and lower prompt information of the main driver, and the tactile prompt information includes but is not limited to vibration information, such as vibration information of the main driver and the steering wheel, to solve the problem that the driver is in a non-driving event such as looking at a mobile phone or talking.

[0179] When the lateral visual prompt is provided, the vehicle atmosphere lamp can be used for dynamic prompt according to different functional urgency of the second vehicle approaching, vehicle avoiding, and the system being unable to avoid and requesting the driver to take over. When the second vehicle is approaching, the approaching direction of the second vehicle is marked through the water effect of the vehicle atmosphere lamp according to the relative direction of the second vehicle to the surrounding vehicle, and the approaching distance of the second vehicle can be marked and displayed through color. When the surrounding vehicle is performing avoiding, the atmosphere lamp is displayed in the water effect in the corresponding direction according to the avoiding direction of the surrounding vehicle. When the system requests to take over, the atmosphere lamp is prompted in a red flashing state to enhance the attention degree of the state.

[0180] When the lower visual prompt is on, it includes foot atmosphere light and foot lighting light, which is called based on vehicle configuration; both kinds of foot lights provide flickering state in the process of announcing the second vehicle approaching to improve the attention of the main driver, and when the main driver does not look at the front after a period of time after the first reminder, the haptic prompt information such as seat vibration / safety belt vibration / steering wheel vibration is superimposed to improve the attention of the driver.

[0181] The above prompt process is ended after the driver looks at the front for a prompt period.

[0182] In this embodiment, by prompting the driver in different degrees according to the line of sight direction of the driver of the first vehicle, the attention of the driver to the prompt information can be more significantly increased, and the safety degree in the avoidance process is improved.

[0183] In one exemplary embodiment, the vehicle avoidance method further comprises:

[0184] In the case where it is detected that there are multiple second vehicles in the same lane, with similar vehicle visual feature information, and continuous, in the vehicle range of the first vehicle, the multiple second vehicles are regarded as the same second vehicle.

[0185] Specifically, in the case where it is detected that there are multiple second vehicles in the same lane, with the same or similar vehicle visual feature information, and continuous, in the vehicle range of the first vehicle, the multiple continuous second vehicles at this time are considered to constitute a second vehicle fleet, and in actual application, the second vehicle fleet is a special vehicle fleet such as wedding car fleet, rescue vehicle fleet, and escort vehicle fleet, at this time, the color, brand, and opening double flash state of all second vehicles in the second vehicle fleet are uniform, and the driving information except the position information is also uniform, and when performing vehicle avoidance, the second vehicle fleet is regarded as a whole, that is, the same second vehicle to perform vehicle avoidance operation.

[0186] In one embodiment, in the case where it is detected that there is a second vehicle fleet in the vehicle range of the first vehicle, the [vehicle number, driving information of each vehicle] is output as the second driving information of the second vehicle fleet.

[0187] In this embodiment, by regarding the multiple second vehicles in the same lane, with the same or similar vehicle visual feature information, and continuous, as a whole, the vehicle avoidance method can also be accurately performed for the second vehicle fleet.

[0188] As shown in Figure 5 The following will describe in detail how to avoid the special vehicle by taking the second vehicle as the special vehicle, as follows:

[0189] I. Special vehicle identification

[0190] Detect whether each vehicle to be identified in the vehicle surrounding range of the ego vehicle is a special vehicle through at least one of visual recognition and auditory recognition, and further detect whether there is a special vehicle in the vehicle surrounding range of the ego vehicle.

[0191] In the process of visual recognition, a video stream in the vehicle surrounding range of the first vehicle can be captured by a camera device installed on the ego vehicle, and a video frame image in the vehicle surrounding range of the first vehicle can be obtained from the video stream, and vehicle visual feature information of each vehicle to be identified in the vehicle surrounding range of the first vehicle can be extracted from the video frame image in the vehicle surrounding range of the first vehicle, the vehicle visual feature information including at least one of vehicle structural feature information and vehicle color feature information, the vehicle color feature information including at least one of vehicle body color feature information and alarm light color feature information, and the vehicle structural feature information including at least one of vehicle body text feature information, alarm light structural feature information, and vehicle body logo feature information. Whether each vehicle to be identified in the vehicle surrounding range of the ego vehicle is a special vehicle can be detected through any one of the vehicle visual feature information.

[0192] In the process of auditory recognition, when it is detected that the volume of the environmental audio exceeds a preset volume threshold, it can be determined whether there is running sound of a special vehicle in the environmental sound by analyzing the frequency spectrum of the environmental sound collected in the vehicle surrounding range of the ego vehicle, and in the case where there is running sound of a special vehicle, it is considered that the environmental sound is valid, and at this time, the area where the special vehicle is located is predicted according to the sound pressure of the environmental sound; or, the voice prompt information emitted by the remaining vehicles except the ego vehicle is monitored, and the special vehicle in the vehicle surrounding range of the ego vehicle is detected according to the special vehicle information in the voice prompt information.

[0193] In the process of combining visual recognition and auditory recognition, when a special vehicle is not detected through vehicle visual feature information, the area information of the special vehicle can be predicted through the above-mentioned auditory recognition method first; then, the vehicle trend in the vehicle surrounding range of the ego vehicle is identified, that is, the first lane number of the first vehicle from the current lane to the adjacent lane of the current lane in the target direction of the special vehicle and the second lane number of the first vehicle from the adjacent lane to the current lane are monitored, the lane information of the special vehicle is detected according to the first lane number and the second lane number, and finally, the special vehicle in the vehicle surrounding range of the first vehicle is detected according to the lane information and the area information.

[0194] In addition, after the special vehicle is detected by the vehicle controller of each vehicle, the information of the special vehicle can be sent to the cloud, and through the information sharing mechanism of the cloud, the information of the special vehicle can be sent to other vehicles that may encounter the special vehicle, so that other vehicles that may encounter the special vehicle also avoid the special vehicle.

[0195] II. Identification of driving information of special vehicles

[0196] When a special vehicle is detected within the perimeter of the ego vehicle, the ego vehicle's driving information can be detected visually and aurally, such as the relative position information between the special vehicle and the ego vehicle, the speed of the special vehicle, etc. The relative position information between the special vehicle and the ego vehicle includes relative lane information and relative orientation information between the special vehicle and the ego vehicle. For example, the relative lane information includes whether the special vehicle and the ego vehicle are in the same lane, and the relative orientation information includes whether the special vehicle is in the area ahead of the ego vehicle's driving route. The specific means can refer to the means in the first point above and will not be repeated here.

[0197] 3. Special Vehicles Avoidance:

[0198] First, determine whether the vehicle needs to avoid the special vehicle. Using the aforementioned technical means, the relative position information between the first and second vehicles, the first speed of the first vehicle, and the second speed of the second vehicle are obtained. Based on the relative position information, the first speed, and the second speed of the first vehicle, determine whether the vehicle needs to avoid the special vehicle.

[0199] Among them, if the relative position information indicates that the second vehicle is in an adjacent lane to the lane where the first vehicle is located, there is no need to control the first vehicle to avoid the second vehicle. If the first vehicle needs to change lanes to the lane where the second vehicle is located at this time, it is necessary to reduce speed and change lanes; if the relative position information indicates that the first vehicle and the second vehicle are in the same lane, and the second vehicle is in the front area of ​​the driving route of the first vehicle, and the first speed is higher than the second speed, it is necessary to control the first vehicle to change lanes and overtake the second vehicle; if the relative position information indicates that the first vehicle and the second vehicle are in the same lane, the first vehicle is in the front area of ​​the driving route of the second vehicle, and the first speed is lower than the second speed, it is necessary to control the first vehicle to avoid the second vehicle.

[0200] If the first vehicle needs to be controlled to avoid the second vehicle, it is also necessary to detect the available avoidance space within the vehicle's surrounding range.

[0201] If the available avoidance space meets the avoidance condition of the first vehicle, the first vehicle is controlled to change lanes to avoid the second vehicle.

[0202] If the available avoidance space does not meet the avoidance condition of the first vehicle, and the first speed of the first vehicle is greater than the preset driving speed threshold, the driving speed of the first vehicle is increased, and the timing of whether the available avoidance space meeting the avoidance condition is detected is started. If the available avoidance space meeting the avoidance condition is not detected within the first preset time range, the avoidance message notification control of the first vehicle is triggered, wherein the avoidance message notification control is used to provide avoidance reminder information and push the avoidance reminder information to other vehicles within the vehicle range of the first vehicle, calling on other vehicles to avoid the ego vehicle, so that the ego vehicle detects that the available avoidance space meets the avoidance condition. If the available avoidance space meeting the avoidance condition is detected within the second preset time range, the first vehicle controls the second vehicle to change lanes to avoid, wherein the second preset time range is longer than the first preset time range. If the available avoidance space meeting the avoidance condition is not detected within the second preset time range, the artificial driving prompt information is pushed to the driver terminal of the first vehicle.

[0203] If the available avoidance space does not meet the avoidance condition of the first vehicle, and the first speed of the first vehicle is less than or equal to the preset driving speed threshold, the minimum allowable distance between the first vehicle and other vehicles is obtained, the tilt angle matched based on the minimum allowable distance is controlled, the first vehicle is moved to the adjacent lane, and the tilt avoidance reminder information is pushed to other vehicles within the vehicle range of the first vehicle. For example, the tilt avoidance reminder information can be to call on other vehicles to avoid the ego vehicle and special vehicles, so that the ego vehicle can tilt to avoid.

[0204] Four, state prompting scheme for avoiding special vehicles:

[0205] The line of sight direction of the driver of the first vehicle is detected; when the line of sight direction represents that the driver looks forward, at least one of the first visual prompt information in the forward direction and the auditory prompt information is provided to the first vehicle, such as voice broadcast prompt information and the like, such as rearward approaching vehicle broadcast and special vehicle prompt avoidance broadcast. The first visual prompt information in the forward direction includes but is not limited to vehicle front state instrument screen, head-up display, central control large screen display, etc.; when the line of sight direction represents that the driver does not look forward, at least one of the first visual prompt information in the forward direction, the auditory prompt information, the tactile prompt information, and the second visual prompt information below or beside the vehicle is provided to the first vehicle, such as vibration of the main driver, safety belt, steering wheel, etc. The second visual prompt information below or beside the vehicle includes but is not limited to foot or side ambient light dynamic prompt, etc.

[0206] In addition, the present solution also provides the concept of a special vehicle team, that is, in the case where a plurality of second vehicles in the same lane, with the same or similar visual feature information, and continuous, are detected within the first vehicle's range, the plurality of special vehicles are regarded as a special vehicle team, and the special vehicle team is regarded as a whole, and is substituted into the above vehicle avoidance solution as a larger special vehicle for vehicle avoidance operation.

[0207] The present solution aims at the special vehicle passing scenario in the intelligent driving process, and realizes the determination of the special vehicle within a certain range and the determination of the relative position between the ego vehicle and the special vehicle through vision, hearing and cloud positioning sharing. For the identified special vehicle, the ego vehicle selects a suitable solution for vehicle avoidance according to the vehicle driving state and the vehicle surrounding state, and provides the maximum passing space for the special vehicle. Meanwhile, the ego vehicle also avoids the remaining targets outside the vehicle through the external microphone during the avoidance process. Meanwhile, the vehicle interior identifies the user's line of sight direction, and provides differentiated prompts to the user through the center screen, vehicle state display screen, in-cabin speaker and atmosphere lamp, etc. hardware, to ensure that the user will not be disturbed under the premise of effective prompting. Therefore, the present solution has the following advantages:

[0208] 1. Technical advantage: Based on the existing technical capabilities of the vehicle, the visual recognition technology and the hearing recognition technology are combined to determine whether a special vehicle is detected, to comprehensively improve the system response distance and accuracy of the special vehicle avoidance scenario. At the same time, different vehicle avoidance solutions can be provided for vehicles in different surrounding environment states, to provide the maximum passing possibility for special vehicles;

[0209] 2. User interest: The existing vehicle hardware is applied through software capability combination to improve the convenience of vehicle use. With limited hardware investment, the ability of the vehicle to deal with special vehicles in the assisted driving / autonomous driving state is reduced, and the user's auxiliary attention pressure is reduced;

[0210] 3. Social interest: Promote the comprehensive development of vehicle intelligent application and social application, and improve the driving intelligence of the whole vehicle in response to social special vehicles.

[0211] 4. Enterprise interest: Improve the social responsibility and technological sense of the vehicle, achieve comprehensive empowerment of vehicle intelligence and social participation, and improve brand power.

[0212] It should be understood that although the steps in the flowcharts involved in the above embodiments are shown in sequence according to the arrows, the steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, the execution of the steps is not strictly limited in sequence, and the steps can be executed in other sequences. Moreover, at least some of the steps in the flowcharts involved in the above embodiments can include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of the steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least part of other steps or steps or stages in other steps.

[0213] Based on the same inventive concept, the embodiments of the present application also provide a vehicle avoidance device for implementing the above-mentioned vehicle avoidance method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more vehicle avoidance device embodiments provided below can refer to the limitations of the vehicle avoidance method in the above, which will not be repeated here.

[0214] In one exemplary embodiment, as shown in Figure 6 A vehicle is provided, and the vehicle includes a vehicle avoidance device 2000, which includes: an avoidance vehicle information acquisition module 200, an avoidance judgment module 400, a first avoidance module 600, and a second avoidance module 800, wherein:

[0215] The avoidance vehicle information acquisition module 200 is configured to acquire first driving information of a first vehicle and second driving information of a second vehicle when it is detected that the second vehicle exists in the vehicle range of the first vehicle, wherein the passing priority of the second vehicle is higher than the passing priority of the first vehicle.

[0216] The avoidance judgment module 400 is configured to detect available avoidance space in the vehicle range when it is determined that the first vehicle needs to control the avoidance of the second vehicle according to the first driving information and the second driving information.

[0217] The first avoidance module 600 is configured to control the first vehicle to perform lane change avoidance on the second vehicle if the available avoidance space meets the avoidance condition of the first vehicle.

[0218] The second avoidance module 800 is configured to control the first vehicle to perform inclined avoidance on the second vehicle if the available avoidance space does not meet the avoidance condition.

[0219] In one embodiment, the vehicle avoiding device 2000 further comprises an avoiding judging module, configured to extract the first speed, the first position information and the first lane information of the first vehicle from the first driving information, and extract the second speed, the second position information and the second lane information of the second vehicle from the second driving information; determine the relative position information between the first vehicle and the second vehicle according to the first position information, the first lane information, the second position information and the second lane information; and judge whether the first vehicle needs to avoid the second vehicle according to the relative position information, the first speed and the second speed.

[0220] In one embodiment, the avoiding judging module is further configured to, if the relative position information indicates that the second vehicle is in the adjacent lane of the lane where the first vehicle is located, it is not necessary to control the first vehicle to avoid the second vehicle; if the relative position information indicates that the first vehicle and the second vehicle are in the same lane, the second vehicle is in the front area of the driving route of the first vehicle, and the first speed is higher than the second speed, it is necessary to control the first vehicle to overtake the second vehicle by lane changing; and if the relative position information indicates that the first vehicle and the second vehicle are in the same lane, the first vehicle is in the front area of the driving route of the second vehicle, and the first speed is lower than the second speed, it is necessary to control the first vehicle to avoid the second vehicle.

[0221] In one embodiment, the first driving information comprises the first speed of the first vehicle; and the second avoiding module 800 is further configured to, when the first speed is less than or equal to a preset driving speed threshold, acquire the minimum allowable distance between the first vehicle and other vehicles; and control the first vehicle to move obliquely to the adjacent lane based on the oblique angle matched with the minimum allowable distance.

[0222] In one embodiment, the first driving information comprises the first speed of the first vehicle; and the vehicle avoiding device 2000 further comprises a third avoiding module 800, which is configured to, if the available avoiding space does not satisfy the avoiding condition and the first speed is greater than a preset driving speed threshold, increase the driving speed of the first vehicle, and return to the step of detecting the available avoiding space in the vehicle surrounding range until the detected available avoiding space satisfies the avoiding condition.

[0223] In one embodiment, the third avoidance module 800 is also used to start timing whether an available avoidance space that meets the avoidance condition is detected if the available avoidance space does not meet the avoidance condition and the first speed is greater than a preset driving speed threshold; if no available avoidance space that meets the avoidance condition is detected within the first preset time range, the avoidance message notification control of the first vehicle is triggered, wherein the avoidance message notification control is used to provide avoidance reminder information and push the avoidance reminder information to other vehicles; if the available avoidance space that meets the avoidance condition is detected within the second preset time range, the first vehicle is controlled to change lanes to avoid the second vehicle, wherein the second preset time range is longer than the first preset time range; if no available avoidance space that meets the avoidance condition is detected within the second preset time range, manual driving reminder information is pushed to the driver terminal of the first vehicle.

[0224] In one embodiment, the vehicle avoidance device 2000 also includes an avoidance vehicle judgment module, which is used to detect the vehicle type recognition result of each vehicle to be identified within the vehicle surrounding range of the first vehicle through at least one of visual recognition and auditory recognition methods; based on the vehicle type recognition result of each vehicle to be identified, detect whether there is a second vehicle within the vehicle surrounding range of the first vehicle.

[0225] In one embodiment, the vehicle avoidance judgment module is also used to obtain a video frame image within the vehicle surrounding range of the first vehicle; extract vehicle visual feature information of each vehicle to be identified within the vehicle surrounding range of the first vehicle from the video frame image, and detect the vehicle type recognition result of each vehicle to be identified within the vehicle surrounding range of the first vehicle based on the vehicle visual feature information.

[0226] In one embodiment, the avoidance vehicle judgment module is also used to collect the ambient sound within the vehicle surrounding range of the first vehicle when it is detected that the volume of the ambient audio exceeds a preset volume threshold, and detect the vehicle type recognition result of each vehicle to be identified within the vehicle surrounding range of the first vehicle based on the spectrum of the ambient sound; and / or monitor the voice reminder information issued by the remaining vehicles except the first vehicle, and detect the vehicle type recognition result of each vehicle to be identified within the vehicle surrounding range of the first vehicle based on the voice reminder information.

[0227] In one embodiment, the avoidance vehicle judgment module is further used to predict the area information of the second vehicle through auditory recognition when the second vehicle is not detected through the vehicle visual feature information; monitor the number of first lanes that change from the current lane of the first vehicle to the adjacent lane of the current lane in the target direction of the first vehicle, and the number of second lanes that change from the adjacent lane to the current lane; detect the lane information of the second vehicle based on the number of first lanes and the number of second lanes; and detect the vehicle type recognition result of each vehicle to be identified within the vehicle surrounding range of the first vehicle based on the lane information and the area information.

[0228] In an embodiment, the vehicle avoiding device 2000 further comprises a cloud sharing module, configured to acquire second travel information of a second vehicle sent by a cloud, wherein the cloud is configured to send the second travel information detected by a vehicle controller of a third vehicle to vehicle controllers of a plurality of vehicles within a preset spatial range; and detect whether the second vehicle exists within a vehicle range of the first vehicle according to the second travel information.

[0229] In an embodiment, the vehicle avoiding device 2000 further comprises an avoiding prompting module, configured to detect a line-of-sight direction of a driver of the first vehicle in a process of controlling the first vehicle to perform lane changing avoidance or tilting avoidance on the second vehicle; provide at least one of auditory prompt information and first visual prompt information matched with the line-of-sight direction to the first vehicle when the line-of-sight direction represents that the driver visually looks ahead; and provide at least one of the first visual prompt information, the auditory prompt information, tactile prompt information, and second visual prompt information matched with the line-of-sight direction to the first vehicle when the line-of-sight direction represents that the driver does not visually look ahead.

[0230] In an embodiment, the vehicle avoiding device 2000 further comprises a vehicle fleet identification module, configured to identify a plurality of second vehicles as a same second vehicle in a case where the plurality of second vehicles exist within the vehicle range of the first vehicle, are in a same lane, have similar vehicle visual feature information, and are continuous.

[0231] Each of the above vehicle avoiding devices can be realized by software, hardware, or a combination thereof. Each of the above modules can be embedded in or independent of a processor in a computer device in a hardware form, or stored in a memory in a computer device in a software form, so as to be called and executed by a processor to perform operations corresponding to each of the above modules.

[0232] In an exemplary embodiment, a computer device, which can be a server, is provided, and an internal structure diagram of the computer device can be as shown in FIG. 8. Figure 7As shown in the figure. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through the system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The database of the computer device is used to store the first driving information of the first vehicle and the second driving information of the second vehicle and other data. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with the terminal outside through the network connection. The computer program is executed by the processor to realize a vehicle avoidance method.

[0233] Those skilled in the art can understand that, Figure 7 The structure shown in the figure is a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or less components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0234] In one embodiment, a computer readable storage medium is provided, which stores a computer program. The computer program is executed by the processor to realize the steps in each method embodiment described above.

[0235] In one embodiment, a computer program product is provided, which includes a computer program. The computer program is executed by the processor to realize the steps in each method embodiment described above.

[0236] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), quantum computing-based data processing logic devices, artificial intelligence (AI) processors, and the like.

[0237] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0238] The above embodiments only express several implementation ways of the present application, and the description is specific and detailed, but it should not be understood as a limitation to the patent scope of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, several modifications and improvements can be made, which all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A vehicle avoidance method, characterized in that: The method comprises: When a second vehicle is detected within a range around the first vehicle, first driving information of the first vehicle and second driving information of the second vehicle are acquired, wherein a passing priority of the second vehicle is higher than a passing priority of the first vehicle; detecting available avoidance space within a range around the vehicle when it is determined, based on the first driving information and the second driving information, that the first vehicle needs to be controlled to avoid the second vehicle; If the available avoidance space meets the avoidance condition of the first vehicle, controlling the first vehicle to change lanes to avoid the second vehicle; If the available avoidance space does not satisfy the avoidance condition, the first vehicle is controlled to tilt to avoid the second vehicle.

2. The method according to claim 1, characterized in that The method further comprises: extracting a first speed, a first position information, and a first lane information of the first vehicle from the first driving information, and extracting a second speed, a second position information, and a second lane information of the second vehicle from the second driving information; determining relative position information between the first vehicle and the second vehicle based on the first position information, the first lane information, the second position information, and the second lane information; Whether it is necessary to control the first vehicle to avoid the second vehicle is determined based on the relative position information, the first speed, and the second speed.

3. The method according to claim 2, characterized in that The determining, based on the relative position information, the first speed, and the second speed, whether it is necessary to control the first vehicle to avoid the second vehicle includes: If the relative position information indicates that the second vehicle is in a lane adjacent to the lane in which the first vehicle is located, there is no need to control the first vehicle to avoid the second vehicle; If the relative position information indicates that the first vehicle and the second vehicle are in the same lane, the second vehicle is in an area ahead of the first vehicle's travel path, and the first speed is higher than the second speed, then the first vehicle needs to be controlled to change lanes and overtake the second vehicle; If the relative position information indicates that the first vehicle and the second vehicle are in the same lane, the first vehicle is in the area ahead of the second vehicle's driving route, and the first speed is lower than the second speed, it is necessary to control the first vehicle to avoid the second vehicle.

4. The method according to claim 1, wherein The first driving information includes a first speed of the first vehicle; The controlling the first vehicle to tilt to avoid the second vehicle includes: When the first speed is less than or equal to a preset driving speed threshold, obtaining a minimum allowable distance between the first vehicle and other vehicles; Based on the tilt angle matched with the minimum allowed distance, the first vehicle is controlled to tilt toward an adjacent lane.

5. The method according to claim 1, wherein The first driving information includes a first speed of the first vehicle; the method further includes: If the available avoidance space does not meet the avoidance condition and the first speed is greater than a preset driving speed threshold, the driving speed of the first vehicle is increased, and the process returns to the step of detecting the available avoidance space within the vehicle perimeter until the detected available avoidance space meets the avoidance condition.

6. The method according to claim 5, characterized in that The method further comprises: If the available avoidance space does not satisfy the avoidance condition and the first speed is greater than a preset driving speed threshold, starting a timer to determine whether the available avoidance space that satisfies the avoidance condition is detected; If the available avoidance space is not detected to meet the avoidance condition within a first preset time period, triggering an avoidance message notification control of the first vehicle, wherein the avoidance message notification control is used to provide avoidance reminder information and push the avoidance reminder information to other vehicles; If it is detected that the available avoidance space meets the avoidance condition within a second preset time range, controlling the first vehicle to change lanes to avoid the second vehicle, wherein the second preset time range is longer than the first preset time range; If it is not detected within the second preset time period that the available avoidance space meets the avoidance condition, manual driving prompt information is pushed to the driver terminal of the first vehicle.

7. The method according to claim 1, characterized in that Before acquiring the first driving information of the first vehicle and the second driving information of the second vehicle when the second vehicle is detected to be within the vehicle perimeter of the first vehicle, the method further includes: detecting a vehicle type recognition result of each to-be-identified vehicle within a perimeter of the first vehicle by at least one of visual recognition and auditory recognition; According to the vehicle type identification result of each of the vehicles to be identified, it is detected whether there is a second vehicle within the vehicle perimeter of the first vehicle.

8. The method according to claim 7, characterized in that Detecting the vehicle type recognition results of each to-be-identified vehicle within the perimeter of the first vehicle by visual recognition, including: Acquire a video frame image within a range surrounding the first vehicle; The vehicle visual feature information of each vehicle to be identified within the vehicle surrounding range of the first vehicle is extracted from the video frame image, and the vehicle type recognition result of each vehicle to be identified within the vehicle surrounding range of the first vehicle is detected based on the vehicle visual feature information.

9. The method according to claim 7, characterized in that Detecting the vehicle type identification results of each to-be-identified vehicle within the perimeter of the first vehicle by means of auditory recognition, including: When it is detected that the volume of the ambient audio exceeds a preset volume threshold, collecting the ambient sound within the range of the first vehicle, and detecting the vehicle type recognition result of each to-be-identified vehicle within the range of the first vehicle based on the spectrum of the ambient sound; and / or The voice prompting information emitted by the remaining vehicles except the first vehicle is monitored, and based on the voice prompting information, a vehicle type recognition result of each vehicle to be recognized within a range around the first vehicle is detected.

10. The method according to claim 8, characterized in that Detecting the vehicle type identification results of each to-be-identified vehicle within the perimeter of the first vehicle by visual and auditory identification, including: When the second vehicle is not detected by the vehicle visual feature information, predicting the area where the second vehicle is located by means of auditory recognition; monitoring a first lane number of lanes changing from a current lane in which the first vehicle is located to a lane adjacent to the current lane, and a second lane number of lanes changing from the adjacent lane to the current lane, in a target direction of the first vehicle; detecting lane information of the second vehicle according to the number of the first lane and the number of the second lane; According to the lane information and the area information, a vehicle type recognition result of each to-be-recognized vehicle within a perimeter of the first vehicle is detected.

11. The method according to claim 1, wherein Before acquiring the first driving information of the first vehicle and the second driving information of the second vehicle when the second vehicle is detected to be within the vehicle perimeter of the first vehicle, the method includes: Obtaining second driving information of a second vehicle sent by a cloud, wherein the cloud is used to send the second driving information detected by a vehicle controller of a third vehicle to vehicle controllers of multiple vehicles within a preset spatial range; detecting, based on the second driving information, whether a second vehicle exists within a range surrounding the first vehicle; or, When it is detected that multiple second vehicles are located in the same lane, have similar vehicle visual feature information, and are continuous within the range around the first vehicle, the multiple second vehicles are regarded as the same second vehicle.

12. The method according to claim 1, characterized in that The method further comprises: detecting a sight direction of a driver of the first vehicle during a process of controlling the first vehicle to change lanes or perform a bank avoidance maneuver with respect to the second vehicle; When the sight line direction indicates that the driver is looking forward, providing at least one of auditory prompt information and first visual prompt information matching the sight line direction to the first vehicle; When the sight line direction indicates that the driver is not looking forward, at least one of the first visual prompt information, the auditory prompt information, the tactile prompt information, and second visual prompt information matching the sight line direction is provided to the first vehicle.

13. A vehicle comprising a vehicle avoidance device, characterized in that: The device comprises: an avoidance vehicle information acquisition module, configured to acquire first driving information of the first vehicle and second driving information of the second vehicle when a second vehicle is detected within a perimeter of the first vehicle, wherein the passing priority of the second vehicle is higher than the passing priority of the first vehicle; an avoidance determination module, configured to detect available avoidance space within a range around the first vehicle when it is determined, based on the first driving information and the second driving information, that the first vehicle needs to be controlled to avoid the second vehicle; a first avoidance module, configured to control the first vehicle to change lanes to avoid the second vehicle if the available avoidance space satisfies the avoidance condition of the first vehicle; The second avoidance module is configured to control the first vehicle to perform an inclined avoidance of the second vehicle if the available avoidance space does not satisfy the avoidance condition.

Images