Intelligent deviation processing method and device, equipment and storage medium

Abstract

The invention relates to an intelligent deviation processing method and device, equipment and a storage medium. The method comprises the following steps: acquiring an obstacle determined according to sensing information of mobile equipment and a lane reference line model containing obstacle projection; determining a potential risk obstacle of the mobile device according to current information of the mobile device, the obstacle and the lane reference line model; performing potential risk assessment on the potential risk obstacle according to a preset assessment mode; and controlling the mobile device to execute different intelligent deviation processing according to a result of the potential risk assessment. According to the scheme provided by the invention, potential risks brought by related obstacles in the driving process of the mobile equipment can be evaluated, and intelligent deviation processing is performed according to the potential risks, so that the driving safety is improved.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to an intelligent deviation processing method, device, equipment and storage medium. Background Art

[0002] In the process of autonomous driving of mobile devices such as vehicles, the vehicle generally drives along the center line of the lane on the HD map or the center line of the lane composed of visual lane lines, that is, the vehicle drives in the center. During the driving process, the vehicle generally adopts corresponding deviation decisions for obstacles on the vehicle's driving route to improve the safety of the driving process.

[0003] In the related technology, during the process of autonomous driving, the vehicle generally only makes corresponding deviation decisions based on whether the obstacle interferes with the vehicle's driving route, without considering whether the obstacle will cause a sense of oppression to the driver or the intelligent deviation processing of obstacles adjacent to the vehicle's driving route, thereby posing potential risks to the vehicle's driving safety. Summary of the invention

[0004] In order to solve or partially solve the problems existing in the related technology, the present application provides an intelligent deviation processing method, device, equipment and storage medium, which can evaluate the potential risks brought by relevant obstacles during the driving of mobile equipment, and perform intelligent deviation processing according to the potential risks, thereby improving driving safety.

[0005] The first aspect of the present application provides an intelligent deviation processing method, comprising:

[0006] Obtain obstacles determined according to perception information of the mobile device and a lane reference line model including obstacle projections;

[0007] Determining a potential risk obstacle of the mobile device according to current information of the mobile device, the obstacle and the lane reference line model;

[0008] Performing a potential risk assessment on the potential risk obstacles according to a preset assessment method;

[0009] According to the result of the potential risk assessment, the mobile device is controlled to perform different intelligent deviation processes.

[0010] In one embodiment, the step of performing a potential risk assessment on the potential risk obstacle according to a preset assessment method includes:

[0011] The state machine management is used to determine the lateral risk and / or longitudinal risk of the potential risk obstacle.

[0012] In one embodiment, the determining the lateral risk and / or longitudinal risk of the potential risk obstacle by using a state machine management includes:

[0013] Determine the lateral distance and the longitudinal distance between the potential risk obstacle and the mobile device;

[0014] Determining the lateral risk of the potential risk obstacle based on a comparison between the lateral distance and a preset lateral risk threshold; and / or,

[0015] The longitudinal risk of the potential risk obstacle is determined based on the comparison between the longitudinal distance and a preset longitudinal risk threshold.

[0016] In one embodiment, the determining the lateral risk of the potential risk obstacle according to the comparison between the lateral distance and a preset lateral risk threshold value includes:

[0017] According to different comparison results between the lateral distance and the preset lateral risk threshold, different counting changes are made to the lateral risk count value, the risk approach count value, and the risk distance count value, and the lateral risk of the potential risk obstacle is determined according to the results of the counting changes.

[0018] In one embodiment, determining the longitudinal risk of the potential risk obstacle according to the comparison between the longitudinal distance and a preset longitudinal risk threshold value includes:

[0019] According to different comparison results between the longitudinal distance and a preset longitudinal risk threshold, different counting changes are performed on the longitudinal risk count value, and the longitudinal risk of the potential risk obstacle is determined according to the result of the counting change.

[0020] In one embodiment, controlling the mobile device to perform different intelligent deviation processes according to the result of the potential risk assessment includes:

[0021] According to the result of the potential risk assessment that both the lateral risk count value and the longitudinal risk count value are greater than a preset risk threshold, the mobile device is controlled to execute lane change, lane deviation or deceleration.

[0022] In one embodiment, controlling the mobile device to perform different intelligent deviation processes according to the result of the potential risk assessment includes:

[0023] According to the result of potential risk assessment, if the risk count value is greater than a preset risk threshold and the deviation safety count value is greater than the preset risk threshold, the mobile device is controlled to perform lane change or lane deviation, otherwise the mobile device is controlled to perform lane change, wherein the risk count value is determined according to the lateral risk count value and the longitudinal risk count value.

[0024] In one embodiment, the method further comprises:

[0025] In the case of controlling the mobile device to execute lane change or lane deviation according to the result of potential risk assessment, if it is detected that the long-term risk count value is greater than a preset long-term risk count value threshold, it is updated to execute lane change.

[0026] A second aspect of the present application provides an intelligent deviation processing device, comprising:

[0027] A model acquisition module, used to acquire obstacles determined according to perception information of the mobile device and a lane reference line model including obstacle projections;

[0028] An obstacle determination module, used to determine a potential risk obstacle of the mobile device according to current information of the mobile device, the obstacle and the lane reference line model;

[0029] A risk assessment module, used to perform a potential risk assessment on the potential risk obstacle according to a preset assessment method;

[0030] The intelligent deviation processing module is used to control the mobile device to perform different intelligent deviation processing according to the result of potential risk assessment.

[0031] A third aspect of the present application provides an electronic device, including:

[0032] Processor; and

[0033] The memory stores executable codes thereon, and when the executable codes are executed by the processor, the processor is caused to execute the method as described above.

[0034] A fourth aspect of the present application provides a mobile device, comprising at least one detection device and the electronic device as described above, wherein the detection device collects perception information of the mobile device environment and transmits the perception information to the electronic device.

[0035] A fifth aspect of the present application provides a computer-readable storage medium having executable code stored thereon. When the executable code is executed by a processor of an electronic device, the processor is caused to execute the method as described above.

[0036] The technical solution provided by this application may have the following beneficial effects:

[0037] The technical solution of the present application first obtains obstacles determined according to the perception information of the mobile device and a lane reference line model containing obstacle projections; then, the potential risk obstacles of the mobile device can be determined according to the current information of the mobile device, the obstacles and the lane reference line model; further, the potential risk of the potential risk obstacles is evaluated according to a preset evaluation method; finally, the mobile device can be controlled to perform different intelligent deviation processing according to the results of the potential risk evaluation. Through the above processing, when an obstacle, such as a vehicle in an adjacent lane, approaches the driving route of the mobile device, such as the driving route of the vehicle, the sense of oppression caused by the obstacle to the driver of the vehicle and the potential risks brought by the obstacles adjacent to the driving route of the vehicle can be fully considered. According to the results of the potential risk evaluation, the mobile device, such as the vehicle, can be controlled to perform different intelligent deviation processing, such as lane change, lane deviation or deceleration, so as to eliminate potential risks through measures such as active intelligent deviation, reduce the psychological discomfort of the driver, improve driving safety, and improve the autonomous driving experience.

[0038] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The above and other objects, features and advantages of the present application will become more apparent by describing in more detail exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the present application.

[0040] Figure 1 It is a flowchart of an intelligent deviation processing method shown in an embodiment of the present application;

[0041] Figure 2 is another flow chart of the intelligent deviation processing method shown in the embodiment of the present application;

[0042] Figure 3 It is a schematic diagram of dividing the risk area of ​​the evaluation reference line in the intelligent deviation processing method shown in the embodiment of the present application;

[0043] Figure 4 It is a schematic diagram of a lateral risk state management state machine in an intelligent deviation processing method shown in an embodiment of the present application;

[0044] Figure 5 It is a schematic diagram of a longitudinal risk state management state machine in an intelligent deviation processing method shown in an embodiment of the present application;

[0045] Figure 6 is a schematic diagram of the structure of an intelligent deviation processing device shown in an embodiment of the present application;

[0046] Figure 7It is a schematic diagram of the structure of an electronic device shown in an embodiment of the present application. DETAILED DESCRIPTION

[0047] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0048] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0049] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0050] In the related art, during the process of autonomous driving, the vehicle generally only makes corresponding deviation decisions based on whether the obstacle interferes with the vehicle's driving route, without considering whether the obstacle will cause a sense of oppression to the driver and without considering the deviation processing of obstacles adjacent to the vehicle's driving route, thereby bringing potential risks to the vehicle's driving safety. The present application provides an intelligent deviation processing method, which can evaluate the potential risks brought by relevant obstacles during the driving of a mobile device, and perform intelligent deviation processing according to the potential risks, thereby improving driving safety.

[0051] The technical solution of the embodiment of the present application can be applied to different products, for example, to mobile devices. Mobile devices can be, for example, vehicle devices or robot devices with the following functions: (1) Devices with manned functions, such as family cars, buses, etc. (2) Devices with cargo functions, such as ordinary trucks, vans, trailers, closed trucks, tank trucks, flatbed trucks, container trucks, dump trucks, special structure trucks, etc. (3) Devices with tool functions, such as logistics distribution vehicles, AGVs (Automated Guided Vehicles), patrol cars, cranes, cranes, excavators, bulldozers, forklifts, road rollers, loaders, off-road engineering vehicles, armored engineering vehicles, sewage treatment vehicles, sanitation vehicles, vacuum trucks, floor washing vehicles, watering trucks, sweeping robots, food delivery robots, shopping guide robots, lawn mowers, golf carts, etc. (4) Devices with entertainment functions, such as entertainment vehicles, playground automatic driving devices, balance cars, etc. (5) Devices with special rescue functions, such as fire trucks, ambulances, power repair vehicles, engineering rescue vehicles, etc. For ease of description, the following embodiments are described using the mobile device as an example of an intelligent driving or autonomous driving vehicle.

[0052] The technical solution of the embodiments of the present application is described in detail below with reference to the accompanying drawings.

[0053] Figure 1 It is a flow chart of the intelligent deviation processing method shown in the embodiment of the present application.

[0054] See also Figure 1 , this application method mainly includes:

[0055] S101, obtaining obstacles determined according to perception information of a mobile device and a lane reference line model including obstacle projections.

[0056] The mobile device may be, for example, a vehicle but is not limited thereto.

[0057] The vehicle can use sensors to identify the perception information of the vehicle environment, including the state of the vehicle, lane line information, vehicle information on the lane line, road conditions, intersection information, ramp information, road facilities and sign information, etc. According to the perception information, obstacles on the lane can be determined. The lane reference line model can be determined using relevant road environment modeling technology. In step S101, obstacles determined according to the perception information of the mobile device and lane reference line models containing obstacle projections can be obtained, including lane reference line models that are passable by all vehicles on the current road. Among them, each lane corresponds to a lane reference line model, and the lane reference line is generally the center line of the lane. Among them, when evaluating a lane reference line model, the lane reference line model can be called an evaluation reference line, the lane reference line model on its left can be called a left impact reference line, and the lane reference line model on its right can be called a right impact reference line.

[0058] S102: Determine potential risk obstacles of the mobile device according to current information of the mobile device, obstacles and a lane reference line model.

[0059] According to the lane reference line model corresponding to the evaluation reference line, the potential risk obstacles on the evaluation reference line can be determined. According to the determined evaluation reference line, its left influence reference line and right influence reference line can be determined, and then according to the acquired obstacles and the lane reference line models corresponding to the left influence reference line and the right influence reference line, the potential risk obstacles on the left influence reference line and the right influence reference line of the vehicle can be screened out.

[0060] S103: Perform a potential risk assessment on the potential risk obstacle according to a preset assessment method.

[0061] After determining the potential risk obstacle, the state machine management can be used to determine the lateral risk and / or longitudinal risk of the potential risk obstacle. For example, the lateral distance and longitudinal distance between the potential risk obstacle and the mobile device are determined; the lateral risk of the potential risk obstacle is determined based on the comparison between the lateral distance and the preset lateral risk threshold; and / or the longitudinal risk of the potential risk obstacle is determined based on the comparison between the longitudinal distance and the preset longitudinal risk threshold.

[0062] The lateral risk of the potential risk obstacle is determined based on the comparison between the lateral distance and the preset lateral risk threshold, including: according to different comparison results of the lateral distance and the preset lateral risk threshold, different counting changes are made to the lateral risk count value, the risk approach count value, and the risk distance count value, and the lateral risk of the potential risk obstacle is determined based on the results of the counting changes.

[0063] The longitudinal risk of the potential risk obstacle is determined based on the comparison between the longitudinal distance and the preset longitudinal risk threshold, including: performing different counting changes on the longitudinal risk count value according to different comparison results of the longitudinal distance and the preset longitudinal risk threshold, and determining the longitudinal risk of the potential risk obstacle according to the result of the counting change.

[0064] Different counting changes are performed, including accumulation or decrement.

[0065] S104, controlling the mobile device to execute different intelligent deviation processing according to the result of the potential risk assessment.

[0066] Among them, according to the result of the potential risk assessment, if the lateral risk count value and the longitudinal risk count value are both greater than the preset risk threshold, the mobile device can be controlled to execute lane change, lane deviation or deceleration.

[0067] Among them, according to the result of potential risk assessment, if the risk count value is greater than a preset risk threshold and the deviation safety count value is greater than the preset risk threshold, the mobile device can be controlled to perform lane change or lane deviation, otherwise the mobile device can be controlled to perform lane change, wherein the risk count value is determined according to the lateral risk count value and the longitudinal risk count value.

[0068] In addition, in the case of controlling the mobile device to execute lane change or lane deviation according to the result of potential risk assessment, if it is detected that the long-term risk count value is greater than a preset long-term risk count value threshold, it is updated to execute lane change.

[0069] From this embodiment, it can be found that the technical solution of the present application first obtains obstacles determined according to the perception information of the mobile device and the lane reference line model containing the obstacle projection; then the potential risk obstacles of the mobile device can be determined according to the current information of the mobile device, the obstacles and the lane reference line model; and further the potential risk assessment of the potential risk obstacles is performed according to the preset assessment method; finally, the mobile device can be controlled to perform different intelligent deviation processing according to the results of the potential risk assessment. Through the above processing, when obstacles such as vehicles in adjacent lanes are close to the driving route of the mobile device such as the driving route of the self-vehicle, the sense of oppression caused by the obstacles to the driver of the self-vehicle and the potential risks brought by the obstacles adjacent to the driving route of the self-vehicle can be fully considered. According to the results of the potential risk assessment, the mobile device such as the self-vehicle is controlled to perform different intelligent deviation processing, such as lane change, lane deviation or deceleration, so as to eliminate potential risks through measures such as active intelligent deviation, reduce the psychological discomfort of the driver, improve driving safety, and improve the autonomous driving experience.

[0070] Figure 2It is another flow chart of the intelligent deviation processing method shown in the embodiment of the present application. This embodiment takes the mobile device as a vehicle as an example but is not limited to this. The present application can perform intelligent deviation evaluation on each lane line reference model under the current road, and can evaluate the situation of each lane reference line, and then determine whether the vehicle needs to perform intelligent deviation based on the evaluation result, thereby improving driving safety.

[0071] See also Figure 2 , this application method mainly includes:

[0072] S201, obtaining obstacles determined according to perception information of the vehicle and a lane reference line model including obstacle projections.

[0073] The vehicle can use sensors to identify the perception information of the vehicle environment, including the state of the vehicle, lane line information, vehicle information on the lane line, road conditions, intersection information, ramp information, road facilities and sign information, etc. Lane lines include the current lane, the left lane of the current lane, the right lane of the current lane, etc. Sensors may include, for example: Lidar, Camera, Millimeter Wave Radar, etc. Among them, other vehicles outside the vehicle itself, i.e., outside the vehicle itself, can be regarded as obstacles to the vehicle's travel. Therefore, the vehicle can determine the obstacles on the lane relative to the vehicle itself based on the perception information.

[0074] The present application can also use relevant road environment modeling technology to determine the lane reference line model. For example, a lane reference line model is generated by building a static data model, a dynamic data model, a predictive data model and other steps. Taking the construction of a static data model as an example, the surrounding lane relationship can be determined according to the position of the vehicle, including determining the current lane where the vehicle is located, the left and right lanes, and the relationship between the preceding and succeeding lanes; based on the determined lane relationship, a lane reference line is formed, and the lane reference line is generally the center line of the lane, such as forming a three-lane reference line; the road facilities and sign information are associated with the formed lane reference line to obtain static data; static semantic analysis is performed on the obtained static data, and the analysis results are superimposed on the static data, thereby completing the construction of the entire static data model.

[0075] The present application may use relevant technologies to project the obstacle information output by perception onto the lane reference line model, thereby calculating the SL information (longitudinal distance and lateral distance information) and conflict type of each obstacle relative to the projected lane reference line model. The conflict types may include but are not limited to the following five categories: irrelevant, same reference line, merging, cutting in, close parallel, and crossing.

[0076] This application can process autonomous driving perception information including obstacle type, posture in the UTM coordinate system, speed information in the UTM coordinate system, etc., and convert it to a lane reference line model for representation. Among them, UTM (Universal Transverse Mercator Grid System) coordinates are a type of plane rectangular coordinates. This coordinate grid system and the projection it is based on are widely used in topographic maps, as reference grids for satellite images and natural resource databases, and other applications that require precise positioning. The UTM coordinate system divides the earth into 60 latitude bands, each of which has its own projection coordinate system. In the field of autonomous driving, it is usually necessary to first obtain the origin coordinates of the latitude band where one is located, and then use it as the origin of the current Cartesian coordinate system, and then obtain the UTM coordinates of each position point in the current Cartesian coordinate system. The Cartesian coordinate system is a general term for rectangular coordinate systems and oblique coordinate systems.

[0077] Through the above processing, the lane reference line model information including obstacle projection can be obtained, which includes all lane reference line models that the vehicle can pass under the current road. Among them, when evaluating a lane reference line model, it can be called the evaluation reference line evaluation_line. The lane reference line model on its left can be called the left influence reference line, and the lane reference line model on its right can be called the right influence reference line. Figure 3 As shown, the lane reference line of the current lane where the vehicle is located is used as the evaluation reference line.

[0078] It should be noted that the present application can perform intelligent deviation evaluation on each lane line reference model under the current road, and can calculate the situation of each lane reference line clearly, and then based on the calculation result, consider whether the reference line where the self-vehicle is located needs the self-vehicle to intelligently deviate, whether its left and right influence reference lines can allow intelligent deviation, etc. For example, assuming that there are lane reference line 1 (left), lane reference line 2 (middle), and lane reference line 3 (right), assuming that the current vehicle, i.e., the self-vehicle, is on lane reference line 2. The present application can evaluate each lane reference line model according to the reference line sequence but is not limited thereto. First, lane reference line 1 can be evaluated. At this time, lane reference line 1 is used as an evaluation reference line, and it has no left influence reference line, and lane reference line 2 is its right influence reference line; then lane reference line 2 can be evaluated. At this time, lane reference line 2 is used as an evaluation reference line, lane reference line 1 is its left influence reference line, and lane reference line 3 is its right influence reference line; finally, lane reference line 3 can be evaluated. At this time, lane reference line 3 is used as an evaluation reference line, and lane reference line 2 is its left influence reference line, and there is no right influence reference line, and so on. It should be noted that the reference line models of each lane may be evaluated in a sequence other than the reference line sequence. Regardless of the sequence, the evaluation of the reference line models of each lane may be completed at the end.

[0079] Relevant data can be recorded under each potential risk obstacle, where the relevant data can be described as follows but not limited to:

[0080] Update status update_flag: indicates the update status of the obstacle's current frame data;

[0081] Lateral distance: indicates the shortest lateral distance from the corner point of the obstacle to the evaluation reference line;

[0082] Longitudinal distance: indicates the longitudinal distance from the corner point of the obstacle to the front end of the vehicle on the evaluation reference line;

[0083] Longitudinal risk range lon_range(back,front): indicates the front and back range thresholds of the longitudinal risk based on the evaluation reference line. front_lon_range is the front longitudinal risk threshold, and back_lon_range is the back longitudinal risk threshold.

[0084] Lateral risk range lat_range (near, far): indicates the near and far range thresholds of lateral risk based on the evaluation reference line. near_lat_range is the near lateral risk threshold, and far_lat_range is the far lateral risk threshold.

[0085] Lateral risk type lat_risk_type: When the lateral distance of the potential risk obstacle is at the far lateral risk threshold far_lat_range, it is at a weak risk level lat_risk_1; when it is at the near lateral risk threshold near_lat_range, it is at a strong risk level lat_risk_2; otherwise, it is at lat_risk_0;

[0086] Longitudinal risk type lon_type: When the potential risk obstacle is within the longitudinal risk range lon_range, the risk level is lon_risk_1, and in other cases the level is lon_risk_0;

[0087] Estimated risk time risking_time: represents the time when the vehicle and the potential risk obstacle overlap in the longitudinal space, that is, the parallel time;

[0088] Estimated risk time threshold parallel_time: indicates the threshold of the parallel time between the vehicle and the potential risk obstacle in the longitudinal space;

[0089] Long-term risk time threshold T: the threshold set for long-term risk time;

[0090] Risk approach count value far_to_near_counter: indicates the cumulative number of frames in which the potential risk obstacle approaches from the far lateral risk threshold far_lat_range to the near lateral risk threshold near_lat_range;

[0091] Risk distance count value near_to_far_counter: indicates the cumulative number of frames in which the potential risk obstacle moves away from the near lateral risk threshold near_lat_range to the far lateral risk threshold far_lat_range;

[0092] Lateral risk count value lat_counter: indicates the accumulated number of frames in which the potential risk obstacle is within the lateral risk range lat_range;

[0093] Longitudinal risk count value lon_counter: indicates the cumulative number of frames in which the potential risk obstacle is within the longitudinal risk range lon_range;

[0094] Risk count value risk_counter: indicates the accumulated number of frames in which the potential risk obstacle is within the lateral risk range lat_range and the longitudinal risk range lon_range;

[0095] Long-term risk count value long_risk_counter: indicates that the potential risk obstacle is within the lateral risk range lat_range and the longitudinal risk range lon_range, and the number of frames in which the parallel time (estimated risk time risking_time) exceeds the estimated risk time threshold parallel_time;

[0096] State switching threshold: indicates the threshold at which the risk state needs to be switched;

[0097] Risk threshold threshold_risk: indicates the threshold set for risk;

[0098] Deviation safety count value offset_safe_counter: indicates the accumulation of risks that can be removed by deviation within the lane from the potential risk obstacle at the far lateral risk threshold far_lat_range;

[0099] Safety behavior safe_action: indicates the operation that the vehicle needs to perform to eliminate the risk, including lane changing, lane deviation, or lane changing, lane deviation and deceleration.

[0100] S202, screening out potential risk obstacles on the left influence reference line and the right influence reference line of the vehicle according to the current information of the vehicle, obstacles and lane reference line model.

[0101] According to the lane reference line model corresponding to the evaluation reference line, the potential risk obstacles on the evaluation reference line can be determined. According to the determined evaluation reference line, its left influence reference line and right influence reference line can be determined, and then according to the acquired obstacles and the lane reference line models corresponding to the left influence reference line and the right influence reference line, the potential risk obstacles on the left influence reference line and the right influence reference line of the vehicle can be screened out.

[0102] For the evaluation reference line, potential risk obstacles are screened from the perceived input obstacles to establish the potential risk obstacle set risk_obstacles for this evaluation reference line. The screening principle for screening potential risk obstacles is the obstacles in the evaluation reference line and its left and right impact reference line space. L and R represent the left and right impact reference lines. For example, L_risk_obstacles is the potential risk obstacle set on the left impact reference line, and R_risk_obstacles is the potential risk obstacle set on the right impact reference line.

[0103] In this application, the update status update_flag of each obstacle in the potential risk obstacle set Risk_obstacles can be set to not updated, which is convenient for subsequent state machine management. Taking into account the accuracy error of the perceived obstacle position information, two lateral distance thresholds can be set. At the same time, combined with the width occupied by the vehicle after expansion, the lateral position of the evaluation reference line can be divided, focusing on the range of the lateral risk range lat_range, including the near lateral risk threshold near_lat_range and the far lateral risk threshold far_lat_range; setting the range of the longitudinal risk range lon_range, including the front longitudinal risk threshold front_lon_range and the back longitudinal risk threshold back_lon_range.

[0104] Among them, the longitudinal risk range lon_range can be determined according to the vehicle speed. For example, the longitudinal risk range lon_range(back, front) can be set to (vehicle front position, 1.2 times vehicle speed position) but is not limited to this, that is, the rear longitudinal risk threshold back_lon_range can be the vehicle front position, and the front longitudinal risk threshold front_lon_range can be 1.2 times vehicle speed position.

[0105] Among them, the lateral risk range can be determined according to the standard lane width of 3.75m and the width of common vehicle models. For example, the lateral risk range lat_range(near,far) can be set to (0.6, 1.0), in meters, that is, the near lateral risk threshold near_lat_range can be 0.6 meters, and the far lateral risk threshold far_lat_range can be 1.0 meters. It should be noted that large vehicles such as commercial vehicles have a higher degree of psychological impact on drivers, so the lateral risk range lat_range(near,far) of large vehicles such as commercial vehicles can be set to (0.8, 1.2), in meters, that is, the near lateral risk threshold near_lat_range can be 0.8 meters, and the far lateral risk threshold far_lat_range can be 1.2 meters.

[0106] like Figure 3 As shown, the potential risk obstacle set R_Risk_obstacles that affects the right reference line of the evaluation reference line evaluation_line_ includes Car_1 and Car_2, but not Car_3.

[0107] S203, using a state machine to manage the lateral risk and longitudinal risk of potential risk obstacles, and determining the result of the potential risk assessment.

[0108] The step S203 may include:

[0109] S2031, determining the lateral distance and longitudinal distance between the potential risk obstacle and the vehicle;

[0110] Among them, the lateral distance between the potential risk obstacle and the vehicle can be the shortest lateral distance from the corner point of the potential risk obstacle to the evaluation reference line of the vehicle, and the longitudinal distance between the potential risk obstacle and the vehicle can be the longitudinal distance from the corner point of the potential risk obstacle on the evaluation reference line to the front end of the vehicle.

[0111] The horizontal distance and the vertical distance can be calculated in the following way:

[0112] The Cartesian coordinate system is converted into the Frenet coordinate system. After the conversion, the coordinates (s, l) in the Frenet coordinate system are obtained according to the UTM coordinates of the corner points of the potential risk obstacles.

[0113] Using the vehicle information, the same method as above is used to calculate the coordinates of the corner point of the vehicle on the side of the potential risk obstacle in the Frenet coordinate system (s_car, l_car);

[0114] Then, the difference between the absolute value of l and l_car with the smallest absolute value can be taken as the lateral distance lateral, and the corresponding corner point s can be taken as the longitudinal distance longitudinal.

[0115] S2032: Determine the lateral risk of the potential risk obstacle based on a comparison between the lateral distance and a preset lateral risk threshold.

[0116] According to the different comparison results between the lateral distance and the preset lateral risk threshold, different counting changes can be made to the lateral risk count value, the risk approach count value, and the risk distance count value, and the lateral risk of the potential risk obstacle can be determined according to the result of the counting change. The different counting changes include accumulation or decrement.

[0117] The following uses the example of the three risk states in the horizontal risk dynamic assessment, namely, no risk state, weak risk state and strong risk state, to explain the horizontal risk dynamic assessment process.

[0118] Among them, lateral is the lateral distance, near_lat_range is the near lateral risk threshold, far_lat_range is the far lateral risk threshold, lat_risk_0 is the risk-free state, lat_risk_1 is the weak risk state, lat_risk_2 is the strong risk state, lat_counter is the lateral risk count value, far_to_near_counter is the risk approach count value, near_to_far_counter is the risk far count value, and threshold is the state switching threshold. The value of the state switching threshold threshold can be set to 5, but is not limited thereto, and can be 4 or 6, etc.

[0119] Among them, when the lateral distance is less than the near lateral risk threshold, it means that the lateral risk of the potential risk obstacle relative to the vehicle's driving route is large; when the lateral distance is greater than the far lateral risk threshold, it means that the lateral risk of the potential risk obstacle relative to the vehicle's driving route is small; when the lateral distance is between the near lateral risk threshold and the far lateral risk threshold, it means that the lateral risk of the potential risk obstacle relative to the vehicle's driving route is medium.

[0120] like Figure 4 As shown, the horizontal risk state management state machine includes:

[0121] Taking the lateral distance as input,

[0122] 1) If there is no historical status or it is in the risk-free state lat_risk_0,

[0123] When the lateral distance lateral < the near lateral risk threshold near_lat_range, it enters the strong risk state lat_risk_2;

[0124] When the lateral distance lateral>=near lateral risk threshold near_lat_range and the lateral distance lateral<far lateral risk threshold far_lat_range, enter the weak risk state lat_risk_1;

[0125] When all the conditions are not met, each counter (lateral risk count value lat_counter, risk approach count value far_to_near_counter, risk far away count value near_to_far_counter) is cleared.

[0126] 2) If the historical status is a weak risk status lat_risk_1,

[0127] When the lateral distance lateral < the near lateral risk threshold near_lat_range, the lateral risk count value lat_counter is accumulated, and the risk approach count value far_to_near_counter is accumulated;

[0128] When the lateral distance lateral>=near lateral risk threshold near_lat_range and the lateral distance lateral<far lateral risk threshold far_lat_range, the lateral risk count value lat_counter is accumulated and the risk approach count value far_to_near_counter is cleared;

[0129] When the lateral distance lateral>=far lateral risk threshold far_lat_range, the lateral risk count value lat_counter decreases, and the risk approach count value far_to_near_counter is cleared;

[0130] When the risk approach count value far_to_near_counter>=state switching threshold threshold, the state is switched to the strong risk state lat_risk_2; when the lateral risk count value lat_counter<=0, the state is switched to the risk-free state lat_risk_0.

[0131] 3) If the historical status is a strong risk status lat_risk_2,

[0132] When the lateral distance lateral < the near lateral risk threshold near_lat_range, the lateral risk count value lat_counter is accumulated, and the risk far count value near_to_far_counterr is cleared;

[0133] When the lateral distance lateral>=near lateral risk threshold near_lat_range and the lateral distance lateral<far lateral risk threshold far_lat_range, the lateral risk count value lat_counter is accumulated and the risk distance count value near_to_far_counter is accumulated;

[0134] When the lateral distance lateral>=far lateral risk threshold far_lat_range, the lateral risk count value lat_counter decreases, and the risk distance count value near_to_far_counter is cleared;

[0135] When the risk distance count value near_to_far_counter>=state switching threshold threshold, the state is switched to the weak risk state lat_risk_1, and when the lateral risk count value lat_counter<=0, the state is switched to the risk-free state lat_risk_0.

[0136] S2033: Determine the longitudinal risk of the potential risk obstacle based on a comparison between the longitudinal distance and a preset longitudinal risk threshold.

[0137] According to different comparison results between the longitudinal distance and the preset longitudinal risk threshold, different counting changes can be made to the longitudinal risk count value, and the longitudinal risk of the potential risk obstacle can be determined according to the result of the counting change. The different counting changes include accumulation or decrement.

[0138] The following uses the example of the two risk states in the vertical risk dynamic assessment, namely, a risk-free state and a risk state, to explain the vertical risk dynamic assessment process.

[0139] Among them, longitudinal is the longitudinal distance, front_lon_range is the front longitudinal risk threshold, back_lon_range is the back longitudinal risk threshold, lon_risk_0 is the risk-free state, lon_risk_1 is the risk state, and lon_counter is the longitudinal risk count.

[0140] Among them, when the longitudinal distance is in the risk range lon_range, it means that the potential risk obstacle is risky relative to the vehicle's driving route.

[0141] like Figure 5 As shown, the longitudinal risk state management state machine includes:

[0142] Taking the longitudinal distance longitudinal as input,

[0143] 1) If there is no historical state or it is in the risk-free state lon_risk_0,

[0144] When the longitudinal distance longitudinal>=back_lon_range and the longitudinal distance longitudinal<front_lon_range, the risk state lon_risk_1 is entered.

[0145] When the above conditions are not met, each count (longitudinal risk count value lon_counter) is cleared;

[0146] 2) If the historical status is the risk status lon_risk_1,

[0147] When the longitudinal distance longitudinal>=back longitudinal risk threshold back_lon_range and the longitudinal distance longitudinal<front longitudinal risk threshold front_lon_range, the longitudinal risk count value lon_counter is accumulated, otherwise it is decreased;

[0148] When the longitudinal risk count value lon_counter<=0, the state is switched to the risk-free state lon_risk_0.

[0149] S204, determining the risk time of the potential risk obstacle on the left impact reference line and the right impact reference line.

[0150] Calculate the time that the potential risk obstacle is within the longitudinal risk range lon_range. The calculation can be based on a uniform speed model or a uniform acceleration model. The embodiment of the present application uses a uniform speed model to calculate the estimated risk time risking_time that the potential risk obstacle is within the longitudinal risk range lon_range. The estimated risk time risking_time represents the time that the vehicle and the potential risk obstacle overlap in the longitudinal space, that is, the parallel time. It should be noted that the estimated risk time risking_time is generally a positive value, that is, it does not matter whether the vehicle is overtaking other vehicles or being overtaken by other vehicles.

[0151] This application can set a long-term risk time threshold T. When the estimated risk time risking_time>long-term risk time threshold T, the long-term risk count value long_risk_counter is accumulated and the risk count value risk_counter is accumulated; when risking_time<=T, the risk count value risk_counter is accumulated and the long-term risk count value long_risk_counter is decreased.

[0152] S205, referring to the risk time of the potential risk obstacle on the left impact reference line and the right impact reference line, and performing different intelligent deviation processing according to the result of the potential risk assessment.

[0153] Among them, executing different intelligent deviation processing may include but is not limited to executing lane changing, lane deviation or deceleration of the vehicle.

[0154] It should be noted that lane changing behavior refers to the vehicle changing from the current lane to another adjacent lane, lane deviation behavior refers to the vehicle deviation in the current lane in the opposite direction of an obstacle determined to have potential risks, and deceleration behavior refers to the vehicle maintaining the current lane and decelerating to keep the vehicle away from the obstacle or following the obstacle at a preset distance.

[0155] After the above calculations, the information of the potential risk obstacle set Risk_obstacles is updated, and the vehicle behavior is evaluated based on the lateral risk count value lat_counter and the longitudinal risk count value lon_counter of the potential risk obstacles.

[0156] When the lateral risk count value lat_counter and the longitudinal risk count value lon_counter are both greater than the risk threshold threshold_risk, it is determined that the potential risk obstacle poses a potential risk to the vehicle, and the vehicle needs to perform intelligent deviation and execute safe behavior safe_action, including lane change, lane deviation or deceleration. Among them, the risk threshold threshold_risk can be set to 5 but is not limited to this, and can also be other values, such as 4 or 6. The value of the risk threshold threshold_risk can be determined according to the node operation frequency. For example, in order to reduce the delay caused by multi-frame cumulative judgment, the judgment time can be less than 0.5s, and the value is determined according to the risk threshold threshold_risk<=0.5*node operation frequency.

[0157] When the risk count value risk_counter> the risk threshold threshold_risk, calculate the maximum deviation max_offset of the vehicle in the opposite direction of the potential risk obstacle without crossing the lane line. Figure 3 As shown. At this time, the risk threshold threshold_risk can be set to 5 but is not limited thereto, and can also be other values, such as 4 or 6. If the maximum deviation in the opposite direction of the potential risk obstacle max_offset + lateral distance lateral> far lateral risk threshold far_lat_range, it is considered that the vehicle can escape the lateral risk range after deviating in the current driving lane, and the deviation safety count value offset_safe_counter is accumulated;

[0158] When the deviation safety count value offset_safe_counter>the risk threshold threshold_risk, the vehicle's safety behavior safe_action is determined to be executing a lane deviation behavior or a lane change behavior, otherwise the safety behavior safe_action is determined to be executing a lane change behavior.

[0159] For a potential risk obstacle for which the safe action safe_action is determined to execute lane deviation or lane change, if its long-term risk count value long_risk_counter> the long-term risk count value threshold, the safe action safe_action is updated to execute lane change, otherwise the execution of lane deviation or lane change is still maintained. The long-term risk count value threshold can be set to 5 but is not limited thereto, and can also be other values, such as 4 or 6. The value of the long-term risk count value threshold can also be determined according to the node operation frequency.

[0160] It should be noted that the feasibility of lane changing behavior can be further evaluated. Generally, lane changing behavior cannot violate the navigation lane changing principle and lane speed advantage principle. For example, when the ego vehicle is about to exit the ramp, it cannot change lanes in the opposite direction due to intelligent deviation; the ego vehicle cannot change lanes to a lane whose average flow speed is much lower than the speed of the ego vehicle's lane.

[0161] Furthermore, for lane changing behavior, the safety of lane changing behavior can be evaluated. When there is no feasible lane changing space, the safe action safe_action is updated to lane deviation or lane deviation during lane changing, and the safe action safe_action is updated to speed reduction during lane changing.

[0162] To summarize, after the above steps, the safe behavior safe_action of each lane reference line model can be obtained, that is, the safe behavior safe_action of each evaluation reference line can be obtained. By executing the safe behavior safe_action to realize the intelligent deviation function of the vehicle, driving safety can be improved.

[0163] The solution adopted in this application is that when a commercial vehicle in the adjacent lane or a passenger car on the line poses a potential risk to the driving safety of the vehicle, the vehicle can actively and intelligently deviate or slow down to eliminate the potential risk and improve driving safety. This application adopts a state machine to manage the lateral risk, longitudinal risk and other states of obstacles, and avoids the jitter of single-frame results caused by perception accuracy through counting, thereby also improving functional stability.

[0164] Corresponding to the aforementioned application function implementation method embodiment, the present application also provides an intelligent deviation processing device, an electronic device and corresponding embodiments.

[0165] Figure 6 It is a structural schematic diagram of an intelligent deviation processing device shown in an embodiment of the present application.

[0166] See also Figure 6 The intelligent deviation processing device 60 of the present application includes: a model acquisition module 61, an obstacle determination module 62, a risk assessment module 63, and an intelligent deviation processing module 64.

[0167] The model acquisition module 61 is used to obtain obstacles determined according to the perception information of the mobile device and a lane reference line model including obstacle projections. The mobile device may be, for example, a vehicle but is not limited thereto. The vehicle may use sensors to identify perception information of the vehicle environment, including the state of the vehicle, lane line information, vehicle information on the lane line, road conditions, intersection information, ramp information, road facilities and sign information, etc. Obstacles on the lane may be determined based on the perception information. The lane reference line model may be determined using relevant road environment modeling technology. The model acquisition module 61 may obtain obstacles determined according to the perception information of the mobile device and a lane reference line model including obstacle projections, including lane reference line models of all passable vehicles on the current road.

[0168] The obstacle determination module 62 is used to determine the potential risk obstacles of the mobile device based on the current information of the mobile device, the obstacles and the lane reference line model. The obstacle determination module 62 can determine the potential risk obstacles on the evaluation reference line based on the lane reference line model corresponding to the evaluation reference line. Based on the determined evaluation reference line, its left influence reference line and right influence reference line can be determined, and then based on the acquired obstacles and the lane reference line models corresponding to the left influence reference line and the right influence reference line, the potential risk obstacles on the left influence reference line and the right influence reference line of the vehicle can be screened out.

[0169] The risk assessment module 63 is used to perform a potential risk assessment on a potential risk obstacle according to a preset assessment method. After determining the potential risk obstacle, the risk assessment module 63 can use a state machine management to determine the lateral risk and / or longitudinal risk of the potential risk obstacle. For example, the lateral distance and longitudinal distance between the potential risk obstacle and the mobile device are determined; the lateral risk of the potential risk obstacle is determined based on the comparison between the lateral distance and the preset lateral risk threshold; and / or the longitudinal risk of the potential risk obstacle is determined based on the comparison between the longitudinal distance and the preset longitudinal risk threshold. The lateral risk of the potential risk obstacle is determined based on the comparison between the lateral distance and the preset lateral risk threshold, including: according to different comparison results between the lateral distance and the preset lateral risk threshold, different counting changes are made to the lateral risk count value, the risk approach count value, and the risk distance count value, and the lateral risk of the potential risk obstacle is determined according to the result of the counting change. The longitudinal risk of the potential risk obstacle is determined based on the comparison between the longitudinal distance and the preset longitudinal risk threshold, including: according to different comparison results between the longitudinal distance and the preset longitudinal risk threshold, different counting changes are made to the longitudinal risk count value, and the longitudinal risk of the potential risk obstacle is determined according to the result of the counting change. Different counting changes are performed, including accumulation or decrement.

[0170] The intelligent deviation processing module 64 is used to control the mobile device to perform different intelligent deviation processing according to the result of the potential risk assessment. Among them, the intelligent deviation processing module 64 can control the mobile device to perform lane change, lane deviation or deceleration according to the result of the potential risk assessment that the lateral risk count value and the longitudinal risk count value are both greater than the preset risk threshold. Among them, the intelligent deviation processing module 64 can control the mobile device to perform lane change or lane deviation according to the result of the potential risk assessment that the risk count value is greater than the preset risk threshold, and the deviation safety count value is greater than the preset risk threshold, otherwise the mobile device is controlled to perform lane change, wherein the risk count value is determined according to the lateral risk count value and the longitudinal risk count value. In addition, in the case where the intelligent deviation processing module 64 controls the mobile device to perform lane change or lane deviation according to the result of the potential risk assessment, if it is detected that the long-term risk count value is greater than the preset long-term risk count value threshold, it is updated to perform lane change.

[0171] It can be found from this embodiment that the technical solution of the present application can fully consider the sense of oppression caused by obstacles to the driver of the own vehicle and the potential risks brought by obstacles adjacent to the driving route of the own vehicle when obstacles such as vehicles in adjacent lanes are close to the driving route of the mobile device such as the driving route of the own vehicle. According to the results of the potential risk assessment, the mobile device such as the own vehicle is controlled to perform different intelligent deviation processing, such as lane changing, lane deviation or deceleration, etc., so as to eliminate potential risks through measures such as active intelligent deviation, reduce the driver's psychological discomfort, improve driving safety, and improve the autonomous driving experience.

[0172] The embodiment of the present application also provides a mobile device, including at least one detection device and an electronic device, wherein the detection device collects point cloud data of the mobile device environment and transmits the point cloud data to the electronic device. The electronic device may be, for example, an AVCU (Automated Vehicle Control Unit) on a vehicle. The electronic device may include the above-mentioned intelligent deviation processing device.

[0173] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated again here.

[0174] Figure 7 The electronic device 1000 of the present application may be an intelligent electronic device applied to a vehicle or a robot.

[0175] See also Figure 7 , the electronic device 1000 includes a memory 1010 and a processor 1020 .

[0176] The processor 1020 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0177] The memory 1010 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. Among them, ROM can store static data or instructions required by the processor 1020 or other modules of the computer. The permanent storage device may be a readable and writable storage device. The permanent storage device may be a non-volatile storage device that does not lose the stored instructions and data even after the computer is powered off. In some embodiments, the permanent storage device uses a large-capacity storage device (such as a magnetic or optical disk, flash memory) as a permanent storage device. In some other embodiments, the permanent storage device may be a removable storage device (such as a floppy disk, optical drive). The system memory may be a readable and writable storage device or a volatile readable and writable storage device, such as a dynamic random access memory. The system memory may store some or all instructions and data required by the processor at run time. In addition, the memory 1010 may include any combination of computer-readable storage media, including various types of semiconductor storage chips (such as DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. In some embodiments, the memory 1010 may include a readable and / or writable removable storage device, such as a laser disc (CD), a read-only digital versatile disc (such as a DVD-ROM, a double-layer DVD-ROM), a read-only Blu-ray disc, an ultra-density optical disc, a flash memory card (such as an SD card, a mini SD card, a Micro-SD card, etc.), a magnetic floppy disk, etc. The computer-readable storage medium does not include carrier waves and transient electronic signals transmitted wirelessly or wired.

[0178] The memory 1010 stores executable codes, and when the executable codes are processed by the processor 1020 , the processor 1020 can execute part or all of the methods described above.

[0179] In addition, the method according to the present application may also be implemented as a computer program or a computer program product, which includes computer program code instructions for executing some or all of the steps in the above method of the present application.

[0180] Alternatively, the present application can also be implemented as a computer-readable storage medium (or non-transitory machine-readable storage medium or machine-readable storage medium) on which executable code (or computer program or computer instruction code) is stored. When the executable code (or computer program or computer instruction code) is executed by a processor of an electronic device (or server, etc.), the processor executes part or all of the steps of the above-mentioned method according to the present application.

[0181] The embodiments of the present application have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. An intelligent deviation processing method, characterized in that: include: Obtain obstacles determined according to perception information of the mobile device and a lane reference line model including obstacle projections; Determining a potential risk obstacle of the mobile device according to current information of the mobile device, the obstacle and the lane reference line model; Performing a potential risk assessment on the potential risk obstacles according to a preset assessment method; According to the result of the potential risk assessment, the mobile device is controlled to perform different intelligent deviation processes.

2. The method according to claim 1, characterized in that The step of performing a potential risk assessment on the potential risk obstacle according to a preset assessment method includes: The state machine management is used to determine the lateral risk and / or longitudinal risk of the potential risk obstacle.

3. The method according to claim 2, characterized in that The adopting state machine management to determine the lateral risk and / or longitudinal risk of the potential risk obstacle includes: Determine the lateral distance and the longitudinal distance between the potential risk obstacle and the mobile device; Determining the lateral risk of the potential risk obstacle based on a comparison between the lateral distance and a preset lateral risk threshold; and / or, The longitudinal risk of the potential risk obstacle is determined based on the comparison between the longitudinal distance and a preset longitudinal risk threshold.

4. The method according to claim 3, characterized in that The determining the lateral risk of the potential risk obstacle according to the comparison between the lateral distance and a preset lateral risk threshold value includes: According to different comparison results between the lateral distance and the preset lateral risk threshold, different counting changes are made to the lateral risk count value, the risk approach count value, and the risk distance count value, and the lateral risk of the potential risk obstacle is determined according to the results of the counting changes.

5. The method according to claim 3, characterized in that: Determining the longitudinal risk of the potential risk obstacle according to the comparison between the longitudinal distance and a preset longitudinal risk threshold includes: According to different comparison results between the longitudinal distance and a preset longitudinal risk threshold, different counting changes are performed on the longitudinal risk count value, and the longitudinal risk of the potential risk obstacle is determined according to the result of the counting change.

6. The method according to claim 1, characterized in that According to the result of the potential risk assessment, controlling the mobile device to perform different intelligent deviation processes includes: According to the result of the potential risk assessment that both the lateral risk count value and the longitudinal risk count value are greater than a preset risk threshold, the mobile device is controlled to execute lane change, lane deviation or deceleration.

7. The method according to claim 1, characterized in that According to the result of the potential risk assessment, controlling the mobile device to perform different intelligent deviation processes includes: According to the result of potential risk assessment, if the risk count value is greater than a preset risk threshold and the deviation safety count value is greater than the preset risk threshold, the mobile device is controlled to perform lane change or lane deviation, otherwise the mobile device is controlled to perform lane change, wherein the risk count value is determined according to the lateral risk count value and the longitudinal risk count value.

8. The method according to claim 7, characterized in that The method further comprises: In the case of controlling the mobile device to execute lane change or lane deviation according to the result of potential risk assessment, if it is detected that the long-term risk count value is greater than a preset long-term risk count value threshold, it is updated to execute lane change.

9. An intelligent deviation processing device, characterized in that: include: A model acquisition module, used to acquire obstacles determined according to perception information of the mobile device and a lane reference line model including obstacle projections; An obstacle determination module, used to determine a potential risk obstacle of the mobile device according to current information of the mobile device, the obstacle and the lane reference line model; A risk assessment module, used to perform a potential risk assessment on the potential risk obstacle according to a preset assessment method; The intelligent deviation processing module is used to control the mobile device to perform different intelligent deviation processing according to the result of potential risk assessment.

10. An electronic device, characterized in that: include: processor; as well as A memory having executable codes stored thereon, which, when executed by the processor, causes the processor to execute the method according to any one of claims 1 to 8.

11. A mobile device, characterized in that: The device comprises at least one detection device and the electronic device as claimed in claim 10, wherein the detection device collects perception information of the environment of the mobile device and transmits the perception information to the electronic device.

12. A computer-readable storage medium having executable codes stored thereon, which, when executed by a processor of an electronic device, causes the processor to execute the method according to any one of claims 1 to 8.

Images