Vehicle escape control method, device, equipment and storage medium

Abstract

The present disclosure relates to a vehicle escape control method, device, equipment and storage medium, and relates to the field of vehicle control technology. The vehicle escape method includes: obtaining the current driving information of the vehicle, and determining the driving state of the vehicle according to the current driving information; when the driving state is a blocked state, determining multiple roads corresponding to the vehicle according to the current driving information; screening out valid roads from multiple roads; calculating the escape position of the vehicle on the valid road; generating an escape trajectory based on the current position and the escape position in the current driving information, and controlling the vehicle to travel based on the escape trajectory. The method provided by the present disclosure enables the unmanned vehicle to break through the safety constraints and traffic environment constraints in complex scenes, and escape from the blocked state to continue to complete the driving task.

Description

Technical Field

[0001] The present disclosure relates to the field of vehicle control technology, and in particular to a vehicle escape control method, device, equipment and storage medium. Background Art

[0002] On the way to complete the task, the autonomous driving vehicle often encounters complex environmental scenes, including narrow areas, U-shaped bend areas, roundabout areas, station parking areas, intersection areas and head-on collisions. In these complex scenes, it is often difficult to ensure that the vehicle continues to maintain good safety and has good traffic capacity. Because these scene environments are complex, it is sometimes difficult for the autonomous driving system of the unmanned vehicle to meet various requirements to continue moving forward, so during the operation of the unmanned vehicle, it is often stuck in these complex scenes. Therefore, in order to improve the operating efficiency of the unmanned vehicle, it is urgently necessary to provide a vehicle escape control method so that the unmanned vehicle can break through the safety constraints and traffic environment constraints in complex scenes, get out of the jammed state and continue to complete the driving task. Summary of the invention

[0003] In order to solve the above technical problems, the present disclosure provides a vehicle escape control method, device, equipment and storage medium, so that unmanned vehicles can break through safety constraints and traffic environment constraints in complex scenarios, escape from congestion and continue to complete driving tasks.

[0004] In a first aspect, an embodiment of the present disclosure provides a vehicle escape control method, comprising:

[0005] Acquiring current driving information of a vehicle, and determining a driving state of the vehicle according to the current driving information;

[0006] In a case where the driving state is a congested state, determining a plurality of roads corresponding to the vehicle according to the current driving information;

[0007] Screening out valid roads from the plurality of roads;

[0008] Calculating the escape position of the vehicle on the valid road;

[0009] An escape trajectory is generated based on the current position in the current driving information and the escape position, and the vehicle is controlled to travel based on the escape trajectory.

[0010] In a second aspect, an embodiment of the present disclosure provides a vehicle escape control device, comprising:

[0011] An acquisition module, used to acquire current driving information of a vehicle and determine a driving state of the vehicle according to the current driving information;

[0012] a determination module, configured to determine, when the driving state is a congested state, a plurality of roads corresponding to the vehicle according to the current driving information;

[0013] A screening module, used for screening out valid roads from the plurality of roads;

[0014] A calculation module, used for calculating the escape position of the vehicle on the effective road;

[0015] A generation module is used to generate an escape trajectory based on the current position in the current driving information and the escape position, and control the vehicle to travel based on the escape trajectory.

[0016] In a third aspect, an embodiment of the present disclosure provides an electronic device, including:

[0017] Memory;

[0018] Processor; and

[0019] Computer programs;

[0020] Wherein, the computer program is stored in the memory and is configured to be executed by the processor to implement the vehicle escape control method as described above.

[0021] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the vehicle escape control method as described above are implemented.

[0022] The embodiment of the present disclosure provides a vehicle escape method, including: obtaining the current driving information of the vehicle, and determining the driving state of the vehicle according to the current driving information; in the case where the driving state is a congested state, determining multiple roads corresponding to the vehicle according to the current driving information; screening out valid roads from the multiple roads, and calculating the escape position of the vehicle on the valid road; generating an escape trajectory based on the current position and the escape position in the current driving information, and controlling the vehicle to travel based on the escape trajectory. The method provided by the present disclosure enables the unmanned vehicle to break through the safety constraints and traffic environment constraints in complex scenarios, escape from the congested state and continue to complete the driving task. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0024] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0025] Figure 1 A schematic diagram of the structure of an automatic driving system provided by an embodiment of the present disclosure;

[0026] Figure 2 A schematic diagram of the structure of a vehicle escape control system provided by an embodiment of the present disclosure;

[0027] Figure 3 A flow chart of a vehicle escape control method provided by an embodiment of the present disclosure;

[0028] Figure 4 A flowchart of a method for implementing S330 provided in an embodiment of the present disclosure;

[0029] Figure 5 A schematic diagram of a driving scenario provided by an embodiment of the present disclosure;

[0030] Figure 6 A flowchart of a method for implementing S340 provided in an embodiment of the present disclosure;

[0031] Figure 7 A schematic diagram of a search starting point on a valid lane provided by an embodiment of the present disclosure;

[0032] Figure 8 A schematic diagram of a vehicle provided in an embodiment of the present disclosure;

[0033] Figure 9 A schematic diagram of information interaction between a vehicle and a cloud provided in an embodiment of the present disclosure;

[0034] Figure 10 A flowchart of a method for implementing S350 provided in an embodiment of the present disclosure;

[0035] Figure 11 A schematic diagram of an exerciseable area provided in an embodiment of the present disclosure;

[0036] Figure 12 A flow chart of a driving state determination method provided by an embodiment of the present disclosure;

[0037] Figure 13 A flow chart of a driving status sending method provided by an embodiment of the present disclosure;

[0038] Figure 14A schematic diagram of a flow chart of a method for determining an escape position provided in an embodiment of the present disclosure;

[0039] Figure 15 A flowchart of an effective lane determination method provided by an embodiment of the present disclosure;

[0040] Figure 16 A schematic diagram of a process for generating an escape trajectory provided by an embodiment of the present disclosure;

[0041] Figure 17 A schematic diagram of the structure of a vehicle escape control device provided by an embodiment of the present disclosure;

[0042] Figure 18 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0043] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0044] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present disclosure, rather than all of the embodiments.

[0045] When autonomous vehicles (such as unmanned vehicles) are completing driving tasks, they often encounter complex environmental scenes, including narrow areas, U-shaped bends, roundabouts, station parking areas, intersections, and head-on collisions. In these complex scenes, if the vehicle can automatically escape from the jam, it means that while ensuring the safety of the vehicle, it must also ensure that the vehicle has good traffic capacity. The problems that need to be solved for the vehicle to automatically escape include at least the following aspects: 1. Identify that the unmanned vehicle is in a jammed environment; 2. Complete escape communication between the vehicle and the cloud; 3. Find the ideal escape location; 4. Calculate the escape trajectory.

[0046] In view of the above technical problems, the disclosed embodiments provide a vehicle escape control method, which is applied to the vehicle side. After the vehicle side identifies that it is in a congested environment, it sends feedback to the cloud server, and then the cloud sends an escape command to the vehicle side, triggering the vehicle side to run the escape control system (hereinafter referred to as the escape system) deployed thereon to determine the escape position and calculate the escape trajectory, so that the vehicle side can break through the safety constraints and traffic environment constraints in complex scenes, escape from the congested environment and continue to complete the driving task. Detailed description is given through one or more of the following embodiments.

[0047] Before describing in detail the vehicle escape control method provided by the embodiment of the present disclosure, it is preferred to describe the automatic driving system and the escape system deployed on the vehicle.

[0048] Figure 1 A structural schematic diagram of an autonomous driving system provided for an embodiment of the present disclosure, wherein a complete autonomous driving system includes an autonomous driving system on the vehicle side and an autonomous driving system on the cloud side, wherein the autonomous driving system on the vehicle side and the autonomous driving system on the cloud side are connected via a network service, wherein the autonomous driving system on the vehicle side includes a high-precision map service, a perception module, a positioning module, a planning module, a control module, a chassis module, a cloud service client, and a vehicle-side process communication service module, wherein the high-precision map service is used to provide data for the perception module, the positioning module, and the planning module, and the perception module, the positioning module, the planning module, the control module, the chassis module, and the cloud service client communicate with the autonomous driving system on the cloud side via the vehicle-side process communication service module. The cloud-based autonomous driving system includes a cloud monitor, a cloud controller, a multi-vehicle collaborative controller and a cloud process communication service module, among which the cloud monitor, the cloud controller and the multi-vehicle collaborative controller communicate with the vehicle-side autonomous driving system through the cloud process communication service module. Specifically, the cloud monitor is used to receive vehicle-side status information, such as vehicle location information, surrounding environment information, and vehicle operating status, and display the above-received information on the display screen of the cloud monitor; the cloud controller is used to send command information to the vehicle-side, such as pull-over parking and route change, and also receives vehicle-side status feedback and service requests. The functions of other modules are not limited here.

[0049] The escape system provided in this application is mainly composed of a vehicle-side computing system and a cloud control system. For example, see Figure 2 , Figure 2 This is a schematic diagram of a vehicle escape control system provided by an embodiment of the present disclosure, wherein the vehicle-side computing system is deployed in the above Figure 1 In the planning module, the vehicle-side computing system includes four computing modules, namely, vehicle jam status recognition module, escape position calculation module, escape trajectory generation module and escape status switching module. The vehicle-side computing system and the cloud control system are connected through the vehicle-cloud communication service. The cloud control system mainly includes the above Figure 1The cloud monitor and cloud controller in the cloud monitor. Among them, the vehicle congestion state identification module is used to determine whether the vehicle is in a congestion state according to the environmental information of the vehicle's driving and the state of the vehicle itself, and send the congestion state to the cloud monitor. After the cloud monitor determines that the vehicle is in a congestion state, it sends an escape instruction to the vehicle side through the cloud controller. Then the escape position calculation module calculates the escape position. After the escape position calculation module is completed, the escape trajectory generation module starts to calculate and sends the calculated trajectory to the cloud monitor. After receiving the trajectory calculated by the vehicle side, the cloud monitor will determine whether the trajectory meets the reasonable, legal, and safe conditions. If it does, it will send an escape confirmation instruction to the vehicle side through the cloud controller. The escape state switching module will always monitor the escape confirmation instruction sent by the cloud. After receiving the escape confirmation instruction, it controls the vehicle to enter the escape state, so that the vehicle starts to automatically escape and get rid of the congestion.

[0050] Figure 3 A flow chart of a vehicle escape control method provided in an embodiment of the present disclosure is applied to the above escape system, specifically including the following steps: Figure 3 The following steps S310 to S350 are shown:

[0051] S310: Acquire current driving information of the vehicle, and determine the driving state of the vehicle according to the current driving information.

[0052] The current driving information includes driving status, driving task, current vehicle speed, planned driving trajectory and environmental information.

[0053] Understandably, the current driving information of the vehicle is obtained, and the driving state of the vehicle is identified based on the current driving information. Among them, the current driving information includes driving state, driving task, current vehicle speed, planned driving trajectory and environmental information. The driving state includes automatic driving state and non-automatic driving state. The automatic driving state may be the state of non-manually controlled vehicles shown by the cloud-controlled vehicle or the automatic driving system controlling the vehicle. The non-automatic driving state refers to the state of manually controlled vehicles. The driving task can be understood as the task completed by the vehicle during driving. The planned driving trajectory refers to the trajectory of the vehicle from the current position to the target position. The driving state of the vehicle includes a congestion state and a non-congestion state. The driving state can be represented by traffic jam. If it is a congestion state, the value of traffic jam is 1, and if it is a non-congestion state, the value of traffic jam is 0.

[0054] Optionally, determining the driving state in S310 may be implemented by the following steps:

[0055] When the driving state is the autonomous driving state, determine whether the driving task is completed; if the driving task is not completed, when the current vehicle speed is less than or equal to the preset vehicle speed, perform obstacle detection based on the planned driving trajectory and the environmental information, and generate an obstacle detection result; when the obstacle detection result indicates that there is an obstacle in front of the vehicle, calculate the target distance between the vehicle and each obstacle in the obstacle detection result; if the target distance is less than the first preset distance, and it is determined that it is allowed to pass in front of the vehicle according to the traffic light of the target lane where the vehicle is located in the environmental information, determine that the vehicle is in a congested state.

[0056] It is understandable that to determine whether the current driving state of the vehicle is an autonomous driving state. If the driving state is not an autonomous driving state and the vehicle may be manually controlled, it can be directly determined that the vehicle is in a non-blocked state. At this time, trafficjam = 0. If the driving state is an autonomous driving state, then it is determined whether the driving task is completed. For example, it is determined whether the current position of the vehicle is the target position. The target position refers to the destination that the vehicle finally wants to reach. If the current position is the target position, it means that the vehicle is in a non-blocked state. At this time, traffic jam = 0. If the current position is not the target position, it means that the vehicle has not completed the driving task yet and needs to continue moving forward. When the current vehicle speed is less than or equal to the preset vehicle speed, preferably, the preset vehicle speed can be 0, that is, the vehicle has not completed the driving task but has stopped moving. In this case, obstacle detection is continued based on the planned driving trajectory and environmental information to generate an obstacle detection result, that is, to determine whether the vehicle cannot move forward due to the presence of obstacles around it. Obstacle detection refers to detecting whether there are obstacles around the vehicle. Among them, the planned driving trajectory is composed of a series of discrete pose points, and each pose point represents a possible movement position of the vehicle. Then, it is checked whether each pose point collides with the obstacles in the environmental information to perform collision detection and generate an obstacle detection result. If a collision occurs, it means that the vehicle will collide with the obstacle if it continues to move according to the planned driving trajectory. Therefore, the vehicle stops moving. If the obstacle detection result is that there are no obstacles around the vehicle, it is determined that the vehicle is in a non-blocked state. At this time, traffic jam = 0. If the obstacle detection result is that there is an obstacle in front of the vehicle during driving, the target distance between the vehicle and each obstacle in the obstacle detection result is continued to be calculated. If the target distance is greater than or equal to the first preset distance, it is determined that the vehicle is in a non-blocked state. If the target distance is less than the first preset distance and it is determined according to the traffic lights in the target lane where the vehicle is located in the environmental information that it is allowed to pass in front of the vehicle, that is, the traffic signal light in the lane where the vehicle is located is not red, in this case, it is determined that the vehicle is in a blocked state. At this time, traffic jam = 1. If the traffic signal light in the lane where the vehicle is located is red, it is determined that the vehicle is in a non-blocked state. At this time, traffic jam = 0. Among them, the first preset distance can be determined by the user's needs and is not limited here. Through the above steps, it is possible to automatically identify whether the vehicle is in a blocked state, excluding special cases to the greatest extent and with a relatively high recognition accuracy.

[0057] Optionally, after determining that the vehicle is in a blocked state, the following steps are specifically included:

[0058] Determine the first moment when the vehicle is in a jammed state; calculate the jammed time of the vehicle from the first moment; if the jammed time is greater than a preset time, send the jammed state to the cloud, so that the cloud generates an escape instruction based on the jammed state, and the escape instruction is used to indicate the calculation of the escape position of the vehicle.

[0059] It is understandable that after determining that the vehicle is in a jam state (traffic jam = 1), the vehicle side will not directly send the jam state to the cloud, but introduce a clock timer to improve the recognition accuracy of the jam state. Specifically, the first moment when the vehicle is in a jam state is obtained, that is, the moment when traffic jam = 1 is calculated. The first moment can also be understood as the jam start time, and the first moment is recorded as T0. The current moment of the vehicle is obtained in real time and recorded as T1. From the first moment, the jam time of the vehicle is calculated according to the current moment, and the jam time is recorded as T2, where T2 = T1-T0. It is determined whether the jam time is greater than the preset time. The preset time can be understood as the jam limit time, which is recorded as T3. Preferably, T3 can be 5s. If the jam time is greater than the preset time, that is, after determining that the vehicle is in a jam state for a period of time, the vehicle side sends the jam state to the cloud again. If the jam time is less than or equal to the preset time, it is re-determined whether the vehicle is still in a jam state. If the vehicle is still in a jam state, the above process is continued. If the vehicle is not in a jam state, T0 is updated to T1. The process of sending the congestion status is that the vehicle congestion status identification module sends the congestion status to the vehicle-side process communication service module, and then the vehicle-side cloud service client obtains the congestion status from the vehicle-side process communication service module. Subsequently, the cloud service client can also send the congestion status to the cloud monitor through the vehicle-to-cloud communication service.

[0060] It is understandable that after the cloud monitor receives the congestion status, the background monitor judges the surrounding environment of the vehicle through the cloud monitor. If the monitor determines that the vehicle is indeed in a congestion state, the vehicle's automatic driving system has fallen into a congestion dilemma and the vehicle cannot continue to move forward according to the planned driving trajectory. At this time, it is necessary to trigger the vehicle-side escape function to enhance the passability of the automatic driving vehicle. The monitor sends an escape command to the vehicle-side through the cloud controller.

[0061] S320: When the driving state is a congested state, determine a plurality of roads corresponding to the vehicle according to the current driving information.

[0062] It is understandable that when it is determined that the driving state is a blocked state, the multiple roads corresponding to the vehicle can be directly determined based on the current driving information. Alternatively, based on the above S310, after the cloud sends an escape instruction to the vehicle, the escape system determines the multiple roads corresponding to the vehicle based on the current driving information, and then uses the escape position calculation module to calculate the escape position. The escape position calculation module is always in a waiting state for receiving the escape instruction sent by the cloud. Receiving the escape instruction indicates that the cloud controller hopes that the vehicle can escape autonomously. Specifically, after the vehicle determines that the driving state is a blocked state and receives the escape instruction, it determines the multiple roads corresponding to the vehicle based on the current driving information. Specifically, the multiple roads can be determined based on environmental information, positioning information, and the current position, among which the multiple roads may be the road where the vehicle is located, the adjacent roads of the road where the vehicle is located, and the roads that are not adjacent to the road where the vehicle is located but allow the vehicle to pass.

[0063] S330: Filter out valid roads from the multiple roads.

[0064] It is understandable that, based on the above S320, the validity of each lane in the multiple lanes is determined according to traffic rules and traffic environment, and the valid lanes are screened out. The valid road refers to the road on which the vehicle can effectively travel, and the number of valid roads can be less than or equal to the number of multiple roads. valid Indicates that if lane valid =0 means the lane is an invalid lane. valid =1 indicates that the lane is a valid lane. If the lane is valid, it is necessary to determine the search starting point on the lane. If the lane is invalid, there is no need to search on the lane.

[0065] In one embodiment, selecting a valid road from the plurality of roads includes:

[0066] For each of the multiple lanes, determine whether the lane is the target lane where the vehicle is located. If so, determine the lane as a valid lane; if not, if the lane is an adjacent lane of the target lane and the lane line of the lane allows passage, and if it is determined that a dynamic obstacle in the lane is in front of the vehicle, determine the lane as a valid lane.

[0067] Specifically, Figure 4 A flowchart of a method for implementing S330 is provided in an embodiment of the present disclosure. There are multiple methods for implementing S330, and the present application does not limit this. The method for implementing S330 of the present invention includes S331 to S332:

[0068] S331. For each of the multiple lanes, determine whether the lane is the target lane where the vehicle is located. If so, determine the lane as a valid lane.

[0069] S332. If not, when the lane is an adjacent lane of the target lane and the lane line of the lane allows passage, and when the dynamic obstacle in the lane is in front of the vehicle during driving, determine the lane as a valid lane.

[0070] It can be understood that for each of the multiple lanes, according to the lane information of the lane, it is determined whether the lane is the target lane where the vehicle is located, that is, it is determined whether the vehicle is driving on the lane. If the vehicle is driving on the lane, the lane is directly determined as a valid lane. If the vehicle is not driving on the lane, it is determined whether the lane is an adjacent lane of the target lane. If the lane is not an adjacent lane of the target lane, the lane is determined as an invalid lane. If the lane is an adjacent lane of the target lane, it is determined whether the lane line of the lane is a solid line, that is, it is determined whether the vehicle can drive from the target lane to the lane. If the lane line of the lane is a solid line, the lane is determined as an invalid lane. If the lane line of the lane is not a solid line, the dynamic obstacles are determined among all the obstacles in the lane, and it is determined whether all the dynamic obstacles on the lane are driving behind the vehicle. If all the dynamic obstacles on the lane are driving behind the vehicle, the lane is determined as an invalid lane. If all the dynamic obstacles on the lane are driving in front of the vehicle, the lane is determined as a valid lane.

[0071] Exemplarily, Figure 5 is a schematic diagram of a driving scenario provided by an embodiment of the present disclosure. Figure 5 It includes 2 thickened road boundaries, lane 0, lane 1, and lane 2. Among them, in front of the vehicle driving in lane 1, there are obstacle 1 and obstacle 2. Among them, obstacle 1 is a static obstacle, and obstacle 2 is a dynamic obstacle. Lane 1 is the target lane where the vehicle is located, and the arrow direction on lane 1 is the driving direction of the vehicle. Therefore, lane 1 is directly determined as a valid lane. Lane 0 is an adjacent lane of lane 1, and the lane line adjacent to lane 1 and lane 0 is a dotted line, indicating that the vehicle is allowed to drive from lane 1 to lane 0. The obstacle 4 on lane 0 is a dynamic obstacle, and obstacle 4 is behind the vehicle during driving. Therefore, lane 0 is an invalid lane. Lane 2 is an adjacent lane of lane 1, and the obstacle 3 on lane 2 is a dynamic obstacle. However, the lane line adjacent to lane 1 and lane 2 is a solid line, indicating that the vehicle is not allowed to drive from lane 1 to lane 2. Therefore, lane 2 is an invalid lane. In the Figure 5 shown driving scenario, through the above method of calculating valid lanes, lane 1 can be screened out as a valid lane among multiple lanes such as lane 0, lane 1, and lane 2. Among them, the validity of the lane is expressed as lanevalid , effective lane lane valid =1, invalid lane valid =0.

[0072] S340, calculating the escape position of the vehicle on the effective road.

[0073] It can be understood that S340 can be executed by the above-mentioned escape position calculation module, which includes three core calculation units, namely a search starting point calculation unit, an escape position search unit, and an optimal escape position selection unit. The above-mentioned three calculation units are used to calculate the search starting point, search for the escape position based on the search starting point, determine whether the escape position is valid, and obtain a valid escape position.

[0074] Among them, the escape position refers to the position where the vehicle can escape from the jam. After determining the effective lane, the escape position of the vehicle on the effective road is calculated, that is, a reasonable position is calculated on the effective road. As long as the vehicle drives to the reasonable position, it can escape from the jam.

[0075] In one embodiment, the calculating the escape position of the vehicle on the valid road includes:

[0076] A search starting point is determined on the effective lane; and an escape position of the vehicle on the effective lane is calculated based on the search starting point and obstacles on the effective lane.

[0077] Specifically, Figure 6 A flowchart of a method for implementing S340 is provided in an embodiment of the present disclosure. There are multiple methods for implementing S340, and the present application does not limit this. The method for implementing S340 of the present invention includes S341 to S342:

[0078] S341. Determine a search starting point on the valid lane.

[0079] S342: Calculate the escape position of the vehicle on the effective lane based on the search starting point and obstacles on the effective lane.

[0080] In one embodiment, determining a search starting point on the valid lane includes:

[0081] All static obstacles on the effective lane are determined, and all static obstacles are sorted according to the distance from the vehicle to generate an obstacle list; multiple vertices of the first obstacle in the obstacle list are determined, and the vertex farthest from the vehicle among the multiple vertices is used as a target vertex; a perpendicular line is drawn from the target vertex to the center line of the effective lane to determine the intersection of the perpendicular line and the center line; and a search starting point is determined on the effective lane according to the intersection.

[0082] It is understandable that for each valid lane, it is necessary to determine the search starting point on the valid lane, and the search starting point is recorded as point start The process of determining the search starting point is as follows: Count all static obstacles on the valid lane. Static obstacles can be vehicles parked on the lane. Specifically, they can be all static obstacles within the preset range of the lane. The preset range can be 200m. Then, all static obstacles are sorted according to the distance from the vehicle to generate an obstacle list. The obstacle list is recorded as list obs Specifically, all static obstacles can be sorted in order from near to far from the vehicle. After obtaining the obstacle list, determine the multiple vertices of the first obstacle in the obstacle list, that is, obtain the first obstacle in the list. Each obstacle has at least one vertex. Calculate the distance from each vertex to the vehicle, and take the vertex farthest from the vehicle among the multiple vertices as the target vertex. The target vertex is recorded as point max Draw a perpendicular line from the target vertex to the center line of the valid lane, and determine the intersection of the perpendicular line and the center line, that is, calculate point max The closest point to the center line of the lane, the closest point (intersection point) is recorded as point close The intersection point can be directly determined as the search starting point, and the search starting point is recorded as point start , that is, point close The corresponding value is stored in point start In, complete point start It is understandable that in order to reduce the amount of subsequent searches for the escape location based on the search starting point, point start Move forward a fourth preset distance, which is recorded as dist0, and use the point after moving the fourth preset distance as the last search starting point on the valid lane start , initialization is completed.

[0083] For example, Figure 7 A schematic diagram of a search starting point on a valid lane provided in an embodiment of the present disclosure, wherein the valid lane is Figure 5 Lane 1 in , which is the target lane where the vehicle is located, such as Figure 7 As shown, obstacle 1 is the static obstacle closest to the vehicle (the first static obstacle or the first static obstacle in the obstacle list), obstacle 2 is a dynamic obstacle, obstacle 1 includes 4 vertices, which are respectively recorded as vertex 1 to vertex 4. Vertex 2, which is the farthest from the vehicle among the 4 vertices, is taken as the target vertex, and then a perpendicular line is drawn from vertex 2 to the center line of the effective lane to determine the intersection of the perpendicular line and the center line, which is recorded as intersection 1, and intersection 1 is determined as the search starting point.

[0084] Furthermore, in order to reduce the amount of subsequent searches for the escape location based on the search starting point, the nearest point (intersection 1) point close Move forward the fourth preset distance dist0 to get the search starting point point start ,Right now Figure 7 The starting point for the search in .

[0085] In one embodiment, calculating the escape position of the vehicle on the valid lane based on the search starting point and the obstacles on the valid lane includes:

[0086] The search points are obtained based on the search starting point, the second preset distance and the third preset distance; a virtual model of the vehicle is constructed for each of the search points; a collision detection is performed between the virtual model and the static obstacles on the effective lane to obtain a collision detection result corresponding to each of the search points; based on the collision detection result corresponding to each of the search points, the position of the search point corresponding to the collision detection result of no collision on the effective lane is determined as the escape position.

[0087] It can be understood that after determining the search starting point on each valid lane, for each valid lane, at least one search point is obtained based on the search starting point, the second preset distance and the third preset distance on the valid lane. Specifically, the search starting point is offset to the left and / or right by the second preset distance to obtain a new search point, and then the new search point is further offset to the left and / or right by the second preset distance to obtain another new search point, and so on, until the latest search point exceeds the lane line of the valid lane, then the search is stopped, and a group of search points is obtained, recorded as the first group of search points, the first group of search points includes at least one search point including the search starting point, wherein the second preset distance is recorded as dist lat It is understandable that after obtaining a group of search points, the validity of the first group of search points can be judged. If the first group of search points are invalid, the first group of search points are shifted forward by the third preset distance along the direction of the lane line to obtain the second group of search points, and then the validity of the second group of search points is judged until valid search points are obtained. It is also possible to calculate multiple groups of search points and determine the validity of each group of search points starting from the first group of search points. The specific method of obtaining the search points is not limited, wherein the third preset distance is recorded as dist1, dist1 can be the resolution of the sampling point, which is the resolution of the effective lane direction, and preferably, dist1 can be 0.2m.

[0088] For example, Figure 7 As shown, based on the search starting point, the second preset distance and the third preset distance, multiple groups of search points are searched and obtained, which are recorded as the first group of search points to the i-th group of search points, and each group includes at least one search point, such as Figure 7The first group of search points selected by the dotted lines in FIG. 1 includes 5 search points, and the specific search process is not described here.

[0089] It is understandable that after obtaining at least one search point, the process of judging the validity of each search point is as follows: construct a virtual model of the vehicle for each search point, that is, expand each search point into the shape of the vehicle, and at the same time, in order to ensure the safety of the escape position, increase the safety distance around the virtual model to obtain an expanded virtual model, wherein the safety distance can be determined according to user needs. First, determine whether each search point in the first group of search points is valid, perform collision detection on the expanded virtual model and the static obstacles on the effective lane, and obtain the collision detection result corresponding to each search point. The collision detection refers to whether the vehicle will collide with the static obstacle during driving. Based on the collision detection result corresponding to each search point, the position of the search point corresponding to the collision detection result of no collision on the effective lane is determined as the escape position, that is, if the expanded virtual model obtained based on the search point does not collide with the static obstacle, then the search point is considered valid, and the position of the search point on the effective lane is used as the escape position. If a collision occurs, it means that the search point is invalid, and the next group of search points is selected, or the validity detection of the next search point is continued.

[0090] For example, see Figure 8 , Figure 8 A schematic diagram of a vehicle provided in an embodiment of the present disclosure, Figure 8 Based on the above Figure 7 A top view of the virtual model constructed at a certain search point in Figure 8 A virtual model including search points and a vehicle shape constructed based on the search points, such as Figure 8 As shown, there are safety distances on the left and right sides of the vehicle, respectively, which are recorded as the safety distance on the left side of the vehicle and the safety distance on the right side of the vehicle. There are also safety distances at the rear and front of the vehicle, which are recorded as the safety distance at the rear of the vehicle and the safety distance in front of the vehicle. That is, a safety distance is added around the vehicle body, which can be understood as an inflated virtual model. In addition, the safety distance added on each side of the vehicle body is not exactly the same, and can be a pre-set fixed value, which can be set according to the application scenario of the vehicle.

[0091] It is understandable that after searching each valid lane, a valid escape position may be found. For example, if there are valid escape positions in all three valid lanes, it is necessary to select an optimal escape position from the three valid escape positions to generate an optimal escape trajectory, or to generate an escape trajectory based on each valid escape position, that is, to generate three escape trajectories for subsequent use. To select an optimal escape position from the three valid escape positions, it is necessary to calculate the current distance between each valid escape position and the vehicle, and take the escape position corresponding to the shortest current distance as the optimal escape position. After the escape position is calculated, if the goal representing the effectiveness of the escape position is valid The value of is 1, indicating that the escape position is valid. Then the escape trajectory generation module will be called to generate an escape trajectory based on the current position and the escape position. If the goal indicating the validity of the escape position is valid The value of is 0, indicating that the escape position is invalid. At this time, the vehicle will send status information that there is no escape position to the cloud.

[0092] It can be understood that by screening out valid lanes from multiple lanes and excluding invalid lanes, it is helpful to reduce the subsequent calculation amount and search amount, and at the same time it can improve the calculation accuracy of the escape position; then, the search starting point is determined on the valid lane according to traffic rules and traffic environment, which can improve the safety of the vehicle driving along the escape trajectory in the escape mode; based on the search starting point, the second preset distance and the third preset distance, at least one search point is obtained, the search point is constructed into a virtual vehicle model with an increased safety distance, the virtual vehicle model is subjected to collision detection with static obstacles on the valid lane, and the position of the search point corresponding to the collision detection result of no collision on the valid lane is determined as the escape position. This method of determining the escape position can quickly and accurately obtain a safe position for the vehicle to escape from the jam, and can also maximize the safety and passability of the vehicle's driving process to the escape position when automatically escaping.

[0093] S350, generating an escape trajectory based on the current position in the current driving information and the escape position, and controlling the vehicle to travel based on the escape trajectory.

[0094] It is understandable that, based on the above embodiments, after the escape position calculation module is completed, the calculation of the escape trajectory generation module is started. Specifically, when the escape position is valid, an escape trajectory is generated based on the current position of the vehicle and the escape position. The escape trajectory refers to the driving trajectory of the vehicle from the current position to the escape position to escape the jam. Specifically, the trajectory generation algorithm can be used to search for the trajectory. It is understandable that if the trajectory is searched out, it means that the trajectory is valid, and the escape trajectory can be directly output, and the effective escape trajectory can be sent to the cloud. If the trajectory is not searched out, it means that it is impossible to continue to move forward in the current environment, and the information of the invalid trajectory is output, and the information of the trajectory generation failure is sent to the cloud. After the cloud monitor receives the escape trajectory calculated by the vehicle end, it will further determine whether the escape trajectory meets the reasonable, legal and safe escape conditions. If the escape trajectory meets the escape conditions, the cloud controller sends an escape confirmation instruction to the vehicle end.

[0095] It is understandable that the escape status switching module on the vehicle side will always monitor the escape confirmation command sent by the cloud. Once the escape confirmation command is received, it will control the vehicle to enter the escape state, drive to the escape position according to the escape trajectory, and complete autonomous escape.

[0096] For example, see Figure 9 , Figure 9 A schematic diagram of information interaction between a vehicle and the cloud provided for an embodiment of the present disclosure, that is, a vehicle-cloud information interaction process during a successful escape from a jam on the vehicle side, the specific vehicle-cloud interaction process is as follows: the vehicle side sends the vehicle jam status to the cloud, the cloud sends an escape instruction to the vehicle side, the vehicle side sends an escape trajectory to the cloud, the cloud sends an escape confirmation instruction to the vehicle side, and the vehicle-cloud information interaction is completed. After receiving the escape confirmation instruction from the cloud, the escape status switching module on the vehicle side will switch from the automatic driving mode to the escape mode. In the escape mode, the vehicle will travel according to the escape trajectory. After the vehicle travels from the current position to the escape position, the vehicle will stop moving, exit the escape mode, switch to the automatic driving mode, and start to continue to perform the driving tasks that the vehicle has not yet completed according to the planned driving trajectory.

[0097] A vehicle escape control method provided by an embodiment of the present disclosure is applied to the above-mentioned escape system, including: obtaining the current driving information of the vehicle, and automatically identifying the driving state of the vehicle according to the current driving information, the driving state including a blocked state and a non-blocked state, and when the driving state is a blocked state, determining multiple roads corresponding to the vehicle according to the current driving information, then screening out valid roads from the multiple roads, and calculating the escape position of the vehicle on the valid roads, generating an escape trajectory based on the current position of the vehicle and the escape position, and controlling the vehicle to travel based on the escape trajectory, and when the vehicle travels to the escape trajectory, the vehicle is free from the jam. The method provided by the present disclosure enables the vehicle to automatically break through the safety constraints and traffic environment constraints in complex scenarios, and escape from the blocked state to continue to complete the driving task.

[0098] Based on the above embodiments, see Figure 10 , Figure 10 A flowchart of a method for implementing S350 provided in an embodiment of the present disclosure, wherein optionally, generating an escape trajectory based on the current position in the current driving information and the escape position specifically includes: Figure 10 The following steps S101 to S104 are shown:

[0099] It is understandable that after the escape position calculation module is completed, the escape trajectory generation module starts to calculate. If the escape position calculated by the escape position calculation module is valid, the escape trajectory generation module will calculate the escape trajectory. If the escape position is invalid, the escape trajectory generation module will stop calculating. The escape trajectory generation module specifically includes three calculation units, namely, a safety constraint generation unit, a drivable area calculation unit and a trajectory generation unit. Among them, the safety constraint generation unit is used to ensure the safe width of the driving trajectory and obstacles. Generally speaking, the greater the distance between the driving trajectory and the obstacle, the better the safety. However, too high safety may also reduce the vehicle's passability; the drivable area calculation unit is used to provide the area where the vehicle can currently travel; the trajectory generation unit is used to generate an escape trajectory from the current position to the escape position based on the safe width and the drivable area. The specific implementation process of the three calculation units, namely the safety constraint generation unit, the drivable area calculation unit and the trajectory generation unit, can be found in the following embodiments.

[0100] S101. Obtain a preset drivable area and a safe width of the vehicle.

[0101] It can be understood that based on the above relationship between safety and passability, this embodiment introduces two levels of safety width, namely, the first safety width and the second safety width. The second safety width is smaller than the first safety width. The introduction of the two-level safety width can maximize the improvement of safety while ensuring passability. Among them, the first safety width is recorded as Preferably, is 0.5 meters, and the second-level safety width is recorded as Preferably, The safety width is 0.1 meters. The safety width is used to ensure that the vehicle and the obstacle will not collide in practice. Safety can be ensured if the width between the vehicle and the obstacle is greater than the safety width.

[0102] It is understandable that the preset drivable area obtained based on the current position of the vehicle is different in different scenarios. For example, the drivable area in the intersection scenario is the range of the intersection, and the drivable area in the U-turn scenario is the range of the U-turn. Among them, the range of the drivable area is usually larger than the area occupied by the lane, and the lane at the U-turn is smaller than the drivable area. The drivable area is the area where the vehicle can drive. If the vehicle is stuck in the U-turn scenario, it is necessary to use the range of the U-turn to generate an escape trajectory when the vehicle automatically escapes. It is understandable that the drivable area can be drawn in advance and stored in the high-precision map. During driving, the vehicle can use the high-precision map service to read the drivable area in real time based on the current position, which effectively speeds up the generation of the escape trajectory. The drivable area is usually represented by a series of point sets, recorded as space drive ={p0,p1,...,p n Different scenes have corresponding drivable areas. The shape and size of the drivable areas are not fixed and are drawn according to the driving environment.

[0103] For example, see Figure 11 , Figure 11 A schematic diagram of a drivable area provided in an embodiment of the present disclosure, specifically a schematic diagram of a drivable area in a U-shaped bend scene, such as Figure 11 As shown, two lane lines form a lane, and the vehicle travels on the lane. The area in front of the vehicle that is framed by a dotted line is a drivable area.

[0104] S102, calculating a first cost from the current position to the target node in the current driving information, and calculating a second cost from the target node to the escape position.

[0105] The target node is a node searched within the executable area.

[0106] It is understandable that, based on the above S101, a search algorithm is used to search for the target node in the drivable area using the first safety width and the second safety width. The search algorithm can be a hybrid A-star algorithm, and the specific search algorithm used to generate the escape trajectory is not limited. It is understandable that the cost value formula of the hybrid A-star algorithm is f=g+h, where f is the total cost value of a node, g is the actual cost value from the current position of the vehicle to a target node, which is the above-mentioned first cost. The actual cost value can be understood as the distance (curve distance or straight line distance), usually the length of the path; h is the heuristic value from the target node to the escape position, which is the above-mentioned second cost. The heuristic value is usually replaced by the Euclidean distance. For node i, if there is f i =g i +h i , the cost value of a child node q of node i is f q =g i +g(i,q)+q i , g(i,q) refers to the actual cost from node i to child node q.

[0107] S103: Perform collision detection on the target path from the current position to the target node and the safety width to obtain a collision detection result.

[0108] It can be understood that, based on the above S102, the target path from the current position to the target node is subjected to collision detection with the safety width to obtain a collision detection result, that is, to determine whether the vehicle will collide with an obstacle when traveling along the target path.

[0109] S104: When the collision detection result is no collision, if the sum of the first cost and the second cost is less than a preset threshold, generating an escape trajectory according to the current position, the target node and the escape position.

[0110] It can be understood that, based on the above S103, after obtaining the collision-free detection result, the sum of the first cost and the second cost is calculated, that is, the total cost f of the target node is calculated. i =g i +h i If the total cost is less than the preset threshold, an escape trajectory is generated according to the current position, the target node and the escape position. The escape trajectory refers to the driving trajectory starting from the current position, passing through the position of the target node on the valid lane, and reaching the escape position.

[0111] It is understandable that when the hybrid A-star algorithm is used to generate an escape trajectory, the map used is a grid map, so the boundary of the drivable area needs to be converted into a grid map to facilitate the subsequent calculation of the escape trajectory by the hybrid A-star algorithm.

[0112] It is understandable that if the total cost is greater than or equal to the preset threshold, it is necessary to continue searching for multiple child nodes corresponding to the target node until the optimal escape trajectory is generated.

[0113] Among them, the safety width includes a first safety width and a second safety width, the first safety width is greater than the second safety width, and accordingly, collision detection based on different safety widths will generate different collision detection results, the collision detection results include a first collision result corresponding to the first safety width and a second collision result corresponding to the second safety width, no collision includes no first collision corresponding to the first safety width and no second collision corresponding to the second safety width, and collision includes a first collision corresponding to the first safety width and a second collision corresponding to the second safety width.

[0114] Optionally, if the collision detection result obtained in S103 above is that there is a collision, an escape trajectory can be generated in two ways.

[0115] Optionally, the first method is implemented by the following steps:

[0116] If the sum of the first cost and the second cost is greater than or equal to the preset threshold, multiple child nodes corresponding to the target node are searched in the exercisable area; for each of the multiple child nodes, a collision detection is performed on the target path from the target node to the child node with the first safety width to obtain a first collision detection result; if the first collision detection result is no first collision, the child node is determined as a first valid node, and the first valid node is determined as a target valid node; a third cost from the target valid node to the target node is calculated, and a fourth cost from the target valid node to the escape position is calculated; if the sum of the first cost, the third cost and the fourth cost is less than the preset threshold, an escape trajectory is generated according to the current position, the target node, the target valid node and the escape position.

[0117] Understandably, after searching for multiple sub-nodes corresponding to the target node in the drivable area, for each sub-node, the first safety width is preferentially used with the target path from the target node to the sub-node for collision detection to obtain the first collision detection result. If the first collision detection result is that the path from the target node to the sub-node and the first safety width are free of the first collision, it means that the sub-node is valid, and the sub-node is used as the first valid node, and the first valid node is determined as the target valid node. This situation refers to the distance between the vehicle and the obstacle being greater than the first safety width of 0.5m, and the possibility of the vehicle colliding with the obstacle when driving to the position of the sub-node is extremely low. Subsequently, the third cost from the target valid node to the target node, that is, the cost from the sub-node to the target node, the fourth cost from the target valid node to the escape position is calculated, and the sum of the first cost, the third cost and the fourth cost is calculated. If the sum is less than the preset threshold, an escape trajectory is generated according to the current position, the target node, the target valid node and the escape position.

[0118] Optionally, based on the above embodiment, if the first collision detection result is a first collision, that is, the distance between the vehicle and the obstacle is less than the first safety width, it is necessary to continue to perform collision detection with the second safety width. In this case, generating an escape trajectory can be specifically achieved through the following steps: If the first collision detection result is a first collision, the target path from the target node to the child node is subjected to a collision detection with the second safety width to obtain a second collision detection result; if the second collision detection result is no second collision, the child node is determined as a second valid node, and the second valid node is determined as the target valid node; the fifth cost is calculated based on the third cost; if the sum of the first cost, the third cost, the fourth cost and the fifth cost is less than the preset threshold, an escape trajectory is generated based on the current position, the target node, the target valid node and the escape position.

[0119] It can be understood that if the first collision detection result is that there is a first collision, that is, the target path collides with the first safe width, the second safe width is used to continue the collision detection to generate a second collision detection result. If the second collision detection result is that there is no second collision on the target path at this time, the child node is considered valid, and the child node is determined as the second valid node, and the second valid node is determined as the target valid node. In this case, although the child node is valid, the child node is relatively close to the obstacle, so a path penalty cost is added to the child node, recorded as g(i,q), and the total cost of the child node is f q =g i +2g(i,q)+q i, this situation refers to the distance between the vehicle and the obstacle is less than the first safety width of 0.5m and greater than the second safety width of 0.1m, indicating that when the vehicle reaches the position of the child node, there is a certain possibility of collision with the obstacle; if the target path calculated under the second safety width still has a collision, the child node is considered invalid. This situation refers to the distance between the vehicle and the obstacle is less than the second safety width of 0.1m, indicating that when the vehicle reaches the position of the child node, there is a high possibility of collision with the obstacle. Therefore, for safety reasons, the child node can be deleted and the next child node can be searched. By increasing the search method of the safety width, the searched trajectory can be kept away from obstacles, and when the space is relatively narrow, the trajectory can be searched to the maximum extent while ensuring safety.

[0120] Optionally, in a possible scenario, only the first safety width is collided with to obtain a detection result without the first collision, and no collision detection is required with the second safety width, and finally an escape trajectory is generated. In this scenario, the second method can be specifically implemented by the following steps:

[0121] If the sum of the first cost and the second cost is greater than or equal to the preset threshold, multiple child nodes corresponding to the target node are searched in the exercisable area; for each of the multiple child nodes, a collision detection is performed on the target path from the target node to the child node with the first safety width to obtain a first collision detection result, and if the first collision detection result is no first collision, the child node is determined as a target valid node; a third cost from the target valid node to the target node is calculated, and a fourth cost from the target valid node to the escape position is calculated; if the sum of the first cost, the third cost and the fourth cost is less than the preset threshold, an escape trajectory is generated according to the current position, the target node, the target valid node and the escape position.

[0122] Optionally, in another possible scenario, a collision detection is performed with the first safety width to obtain a detection result of a first collision, and then a collision detection is performed with the second safety width to obtain a detection result of no second collision, and finally an escape trajectory is generated. In this scenario, the second method can also be specifically implemented through the following steps:

[0123] If the sum of the first cost and the second cost is greater than or equal to the preset threshold, multiple child nodes corresponding to the target node are searched in the exercisable area; for each of the multiple child nodes, a target path from the target node to the child node is subjected to a collision detection with the first safety width to obtain a first collision detection result; if the first collision detection result is a first collision, a collision detection is performed on the target path with the second safety width to obtain a second collision detection result; if the second collision detection result is no second collision, the child node is determined as a target valid node; a third cost from the target valid node to the target node is calculated, and a fourth cost from the target valid node to the escape position is calculated; a fifth cost is calculated based on the third cost; if the sum of the first cost, the third cost, the fourth cost and the fifth cost is less than the preset threshold, an escape trajectory is generated based on the current position, the target node, the target valid node and the escape position.

[0124] It can be understood that the specific implementation steps of the second method of generating an escape trajectory refer to the first method of generating an escape trajectory mentioned above, and will not be repeated here.

[0125] The disclosed embodiment provides a vehicle escape control method, which searches for a target node in the vehicle's drivable area through a search algorithm, and performs collision detection on the target path from the current position to the target node with the first safety width. If there is no collision on the path, it means that the node is valid, which can ensure the safety and passability of the vehicle from the current position to the position of the target node on the effective lane. If there is a collision on the path, the second safety width and path are used for collision check. If there is no collision on the path at this time, the node is also considered valid, but in this case, the vehicle is relatively close to the obstacle during driving, so the path penalty cost is increased. If the path calculated under the second safety width still has a collision, the node is considered invalid, the node is deleted, and a new node is searched again until a low-cost, high-safety and high-passability escape trajectory is generated. The search method based on the safety width provided by the present disclosure allows the searched trajectory to be away from obstacles, and when the space is relatively narrow, the trajectory can also be searched, thereby improving practicality.

[0126] Based on the above embodiments, see Figure 12 , Figure 12 A flow chart of a driving state determination method provided in an embodiment of the present disclosure, specifically including the following steps: Figure 12 The following steps S121 to S129 are shown:

[0127] S121. Obtain current driving information of the vehicle.

[0128] S122: Determine whether the vehicle is in an automatic driving state.

[0129] Understandable. If so, execute S123; if not, output the recognition result that the vehicle is in a non-blocked state.

[0130] S123. Determine whether the vehicle has completed the driving task.

[0131] Understandable. If not, execute S124; if so, output the recognition result that the vehicle is in a non-blocked state.

[0132] S124. Determine whether the current vehicle speed is greater than 0.

[0133] Understandable. If not, execute S125; if so, output the recognition result that the vehicle is in a non-blocked state.

[0134] S125. Obtain the planned driving trajectory and environmental information output by the planning module for obstacle detection, and generate an obstacle detection result.

[0135] S126. Determine whether there is an obstacle in front of the vehicle according to the obstacle detection result.

[0136] Understandable. If so, execute S127; if not, output the recognition result that the vehicle is in a non-blocked state.

[0137] S127. Calculate the target distance between the vehicle and each obstacle in the obstacle detection result.

[0138] S128. Determine whether the target distance is greater than the first preset distance.

[0139] Understandable. If not, execute S129; if so, output the recognition result that the vehicle is in a non-blocked state.

[0140] S129. Determine whether the traffic light in the target lane where the vehicle is located in the environmental information indicates a red light in front of the vehicle.

[0141] Understandable. If not, output the recognition result that the vehicle is in a blocked state; if so, output the recognition result that the vehicle is in a non-blocked state. Understandable. For the specific implementation steps of S121 to S129 above, refer to the above embodiments and will not be elaborated here.

[0142] Based on the above embodiments, refer to Figure 13 , Figure 13 which is a schematic flowchart of a driving state sending method provided by an embodiment of the present disclosure, specifically including the following steps S131 to S136 as shown in Figure 13 :

[0143] S131. Obtain the driving state of the vehicle.

[0144] S132: Determine whether the vehicle is in a congested state.

[0145] It is understandable that if yes, execute S133, if no, execute S136.

[0146] S133, determining the first moment when the vehicle is in a congested state, and calculating the congestion time of the vehicle from the first moment.

[0147] S134: Determine whether the congestion time is greater than a preset time.

[0148] It is understandable that if yes, execute S135, if no, execute S131.

[0149] S135. Send the congestion status to the cloud.

[0150] S136. Set the current time of the vehicle as the first time.

[0151] It is understandable that the specific implementation steps of the above S131 to S136 refer to the above embodiments and are not described in detail here.

[0152] Based on the above embodiments, Figure 14 A flow chart of a method for determining an escape position provided in an embodiment of the present disclosure is applied to the above-mentioned escape position calculation module, specifically including the following steps: Figure 14 The following steps are shown:

[0153] S141. Determine whether an escape instruction is received.

[0154] It is understandable that if yes, execute S142, if no, execute S146.

[0155] S142. Filter out a valid lane from multiple lanes, and determine a search starting point on the valid lane.

[0156] S143: Calculate the escape position of the vehicle on the effective lane based on the search starting point and obstacles on the effective lane.

[0157] S144. Determine whether the escape position is valid.

[0158] It is understandable that if yes, execute S145, if no, execute S146.

[0159] S145. Generate an escape trajectory based on the current position and the escape position.

[0160] S146. Sending status information of no escape location to the cloud.

[0161] It is understandable that the specific implementation steps of determining the escape position from S141 to S146 mentioned above refer to the above embodiments and will not be described in detail here.

[0162] Based on the above embodiments, Figure 15 A flow chart of an effective lane determination method provided by an embodiment of the present disclosure, specifically including the following steps: Figure 15 The following steps S151 to S157 are shown:

[0163] S151. Acquire lane information of the lane.

[0164] S152. Determine whether the lane is the target lane where the vehicle is located according to the lane information.

[0165] It is understandable that if yes, execute S156, if no, execute S153.

[0166] S153. Determine whether the lane is an adjacent lane to the target lane.

[0167] It is understandable that if yes, execute S154, if no, execute S157.

[0168] S154. Determine whether the lane line of the lane is a solid line.

[0169] It is understandable that if yes, execute S157, if no, execute S155.

[0170] S155: Determine whether there is a dynamic obstacle on the lane, and whether the dynamic obstacle is behind the vehicle.

[0171] It is understandable that if yes, execute S157, if no, execute S156.

[0172] S156. The lane is a valid lane.

[0173] S157. The lane is an invalid lane.

[0174] It is understandable that the specific implementation steps of determining the valid lane from S151 to S157 mentioned above refer to the above embodiment and will not be described in detail here.

[0175] Based on the above embodiments, Figure 16 A schematic diagram of a method for generating an escape trajectory provided by an embodiment of the present disclosure, specifically comprising the following steps: Figure 16 The following steps S161 to S165 are shown:

[0176] S161. Obtain an escape position, a drivable area, and a safe width.

[0177] S162: Based on the escape position, the drivable area and the safe width, a trajectory generation algorithm is called to generate an escape trajectory.

[0178] S163. Determine whether the escape trajectory is valid.

[0179] It is understandable that if yes, execute S164, if no, execute S165.

[0180] S164. Send the escape trajectory to the cloud.

[0181] S165. Sending trajectory generation failure information to the cloud.

[0182] It is understandable that the specific implementation steps of generating the escape trajectory from S161 to S165 mentioned above refer to the above embodiment and will not be repeated here.

[0183] Figure 17 The schematic diagram of the structure of a vehicle escape control device provided by an embodiment of the present disclosure is as follows. The vehicle escape control device provided by an embodiment of the present disclosure can execute the processing flow provided by an embodiment of the vehicle escape control method, such as Figure 17 As shown, the vehicle escape control device 1700 includes an acquisition module 1701, a determination module 1702, a screening module 1703, a calculation module 1704 and a generation module 1705, wherein:

[0184] The acquisition module 1701 is used to acquire the current driving information of the vehicle and determine the driving state of the vehicle according to the current driving information;

[0185] A determination module 1702, configured to determine a plurality of roads corresponding to the vehicle according to the current driving information when the driving state is a congested state;

[0186] A screening module 1703, used for screening out valid roads from the plurality of roads;

[0187] A calculation module 1704, used for calculating the escape position of the vehicle on the valid road;

[0188] The generating module 1705 is used to generate an escape trajectory based on the current position in the current driving information and the escape position, and control the vehicle to travel based on the escape trajectory.

[0189] Optionally, the current driving information in device 1700 includes driving status, driving task, current vehicle speed, planned driving trajectory and environmental information.

[0190] Optionally, the acquisition module 1701 is used to:

[0191] When the driving state is an automatic driving state, determine whether the driving task is completed; if the driving task is not completed, then when the current vehicle speed is less than or equal to a preset vehicle speed, perform obstacle detection according to the planned driving trajectory and the environmental information to generate an obstacle detection result; when the obstacle detection result is that there is an obstacle in front of the vehicle, calculate the target distance between the vehicle and each obstacle in the obstacle detection result; if the target distance is less than a first preset distance, and it is determined that the traffic in front of the vehicle is allowed to pass according to the traffic light of the target lane where the vehicle is located in the environmental information, then determine that the vehicle is in a congested state.

[0192] Optionally, the device 1700 is further used for:

[0193] Determine the first moment when the vehicle is in a jammed state; calculate the jammed time of the vehicle from the first moment; if the jammed time is greater than a preset time, send the jammed state to the cloud, so that the cloud generates an escape instruction based on the jammed state, and the escape instruction is used to indicate the calculation of the escape position of the vehicle.

[0194] Optionally, the calculation module 1704 is used to:

[0195] A search starting point is determined on the effective lane; and an escape position of the vehicle on the effective lane is calculated based on the search starting point and obstacles on the effective lane.

[0196] Optionally, the screening module 1703 is used to:

[0197] For each of the multiple lanes, determine whether the lane is the target lane where the vehicle is located. If so, determine the lane as a valid lane; if not, if the lane is an adjacent lane of the target lane and the lane line of the lane allows passage, and if a dynamic obstacle in the lane is determined in front of the vehicle, determine the lane as a valid lane.

[0198] Optionally, the calculation module 1704 is used to:

[0199] All static obstacles on the effective lane are determined, and all static obstacles are sorted according to the distance from the vehicle to generate an obstacle list; multiple vertices of the first obstacle in the obstacle list are determined, and the vertex farthest from the vehicle among the multiple vertices is used as a target vertex; a perpendicular line is drawn from the target vertex to the center line of the effective lane to determine the intersection of the perpendicular line and the center line; and a search starting point is determined on the effective lane according to the intersection.

[0200] Optionally, the calculation module 1704 is used to:

[0201] The search points are obtained based on the search starting point, the second preset distance and the third preset distance; a virtual model of the vehicle is constructed for each of the search points; a collision detection is performed between the virtual model and the static obstacles on the effective lane to obtain a collision detection result corresponding to each of the search points; based on the collision detection result corresponding to each of the search points, the position of the search point corresponding to the collision detection result of no collision on the effective lane is determined as the escape position.

[0202] Optionally, the generating module 1705 is used to:

[0203] Obtain the preset drivable area and safety width of the vehicle; calculate the first cost from the current position to the target node in the current driving information, and calculate the second cost from the target node to the escape position, wherein the target node is a node searched within the drivable area; perform collision detection on the target path from the current position to the target node and the safety width to obtain a collision detection result; when the collision detection result is no collision, if the sum of the first cost and the second cost is less than a preset threshold, generate an escape trajectory according to the current position, the target node and the escape position.

[0204] Optionally, the safety width includes a first safety width, the collision detection result includes a first collision detection result, and the no collision includes no first collision.

[0205] Optionally, the generating module 1705 is further used for:

[0206] If the sum of the first cost and the second cost is greater than or equal to the preset threshold, multiple child nodes corresponding to the target node are searched in the exercisable area; for each of the multiple child nodes, a collision detection is performed on the target path from the target node to the child node with the first safety width to obtain a first collision detection result; if the first collision detection result is no first collision, the child node is determined as a first valid node, and the first valid node is determined as a target valid node; a third cost from the target valid node to the target node is calculated, and a fourth cost from the target valid node to the escape position is calculated; if the sum of the first cost, the third cost and the fourth cost is less than the preset threshold, an escape trajectory is generated according to the current position, the target node, the target valid node and the escape position.

[0207] Optionally, the safety width further includes a second safety width, the second safety width is smaller than the first safety width, the collision detection result further includes a second collision detection result, and the no collision further includes no second collision.

[0208] Optionally, the generation module 1705 is configured to:

[0209] If the first collision detection result indicates a first collision, perform a collision detection on the target path from the target node to the child node with the second safety width to obtain a second collision detection result. If the second collision detection result indicates no second collision, determine the child node as the second valid node and determine the second valid node as the target valid node; calculate a fifth cost based on the third cost; if the sum of the first cost, the third cost, the fourth cost, and the fifth cost is less than the preset threshold, generate an escape trajectory based on the current position, the target node, the target valid node, and the escape position.

[0210] Optionally, the safety width includes a first safety width, the collision detection result includes a first collision detection result, and the no-collision includes no first collision.

[0211] Optionally, the generation module 1705 is configured to:

[0212] If the sum of the first cost and the second cost is greater than or equal to the preset threshold, search for multiple child nodes corresponding to the target node within the feasible driving area; for each child node among the multiple child nodes, perform a collision detection on the target path from the target node to the child node with the first safety width to obtain a first collision detection result. If the first collision detection result indicates no first collision, determine the child node as the target valid node; calculate a third cost from the target valid node to the target node and calculate a fourth cost from the target valid node to the escape position; if the sum of the first cost, the third cost, and the fourth cost is less than the preset threshold, generate an escape trajectory based on the current position, the target node, the target valid node, and the escape position.

[0213] Optionally, the safety width includes a first safety width and a second safety width, the second safety width is less than the first-level safety width, the collision detection result includes a first collision detection result and a second collision detection result, and the no-collision includes no first collision and no second collision.

[0214] Optionally, the generation module 1705 is configured to:

[0215] If the sum of the first cost and the second cost is greater than or equal to the preset threshold, multiple child nodes corresponding to the target node are searched in the exercisable area; for each of the multiple child nodes, a target path from the target node to the child node is subjected to a collision detection with the first safety width to obtain a first collision detection result; if the first collision detection result is a first collision, a collision detection is performed on the target path with the second safety width to obtain a second collision detection result; if the second collision detection result is no second collision, the child node is determined as a target valid node; a third cost from the target valid node to the target node is calculated, and a fourth cost from the target valid node to the escape position is calculated; a fifth cost is calculated based on the third cost; if the sum of the first cost, the third cost, the fourth cost and the fifth cost is less than the preset threshold, an escape trajectory is generated based on the current position, the target node, the target valid node and the escape position.

[0216] Figure 17 The vehicle escape control device of the illustrated embodiment can be used to execute the technical solution of the above-mentioned method embodiment, and its implementation principle and technical effect are similar and will not be repeated here.

[0217] Figure 18 The electronic device provided by the embodiment of the present disclosure can execute the processing flow provided by the above embodiment, such as Figure 18 As shown, the electronic device 1800 includes: a processor 1801, a communication interface 1802 and a memory 1803; wherein the computer program is stored in the memory 1803 and is configured so that the processor 1801 executes the vehicle escape control method as described above.

[0218] In addition, an embodiment of the present disclosure further provides a computer-readable storage medium on which a computer program is stored, and the computer program is executed by a processor to implement the vehicle escape control method described in the above embodiment.

[0219] In addition, an embodiment of the present disclosure also provides a computer program product, which includes a computer program or instructions, and when the computer program or instructions are executed by a processor, the vehicle escape control method as described above is implemented.

[0220] It should be noted that in this document, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or gateway comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or gateway. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or gateway comprising the said element.

[0221] The above are only specific embodiments of the present disclosure, enabling those skilled in the art to understand or implement the present disclosure. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to these embodiments described herein, but rather will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A vehicle escape control method, characterized in that: include: Acquiring current driving information of a vehicle, and determining a driving state of the vehicle according to the current driving information; When the driving state is a congested state, determining a plurality of lanes corresponding to the vehicle according to the current driving information; Screening out valid lanes from the multiple lanes; Calculating an escape position of the vehicle on the valid lane; generating an escape trajectory based on the current position in the current driving information and the escape position, and controlling the vehicle to travel based on the escape trajectory; The step of selecting a valid lane from the plurality of lanes includes: For each lane of the plurality of lanes, determining whether the lane is a target lane where the vehicle is located, and if so, determining the lane as a valid lane; if not, determining the lane as a valid lane when the lane is an adjacent lane of the target lane and the lane line of the lane is passable and determining a dynamic obstacle in the lane in front of the vehicle; The calculating the escape position of the vehicle on the valid lane includes: A search starting point is determined on the effective lane; a search point is obtained based on the search starting point, a second preset distance and a third preset distance; a virtual model of the vehicle is constructed for each of the search points; a collision detection is performed on the virtual model and a static obstacle on the effective lane to obtain a collision detection result corresponding to each of the search points; based on the collision detection result corresponding to each of the search points, a position of the search point corresponding to a collision-free result on the effective lane is determined as an escape position.

2. The method according to claim 1, wherein The current driving information includes a driving state, a driving task, a current vehicle speed, a planned driving trajectory, and environmental information. The determining the driving state of the vehicle according to the current driving information includes: When the driving state is an automatic driving state, determining whether the driving task is completed; If the driving task is not completed, when the current vehicle speed is less than or equal to the preset vehicle speed, obstacle detection is performed according to the planned driving trajectory and the environmental information to generate an obstacle detection result; When the obstacle detection result indicates that there is an obstacle ahead of the vehicle, calculating a target distance between the vehicle and each obstacle in the obstacle detection result; If the target distance is less than the first preset distance, and it is determined that passage is allowed ahead of the vehicle according to the traffic light of the target lane where the vehicle is located in the environmental information, then it is determined that the vehicle is in a congested state.

3. The method according to claim 2, wherein After determining that the vehicle is in a blocked state, the method further includes: Determining a first moment when the vehicle is in a jammed state; Calculate the congestion time of the vehicle from the first moment; If the congestion time is greater than a preset time, the congestion status is sent to the cloud, so that the cloud generates an escape instruction based on the congestion status, and the escape instruction is used to indicate the calculation of the escape position of the vehicle.

4. The method according to claim 1, wherein Determining a search starting point on the valid lane includes: Determine all static obstacles on the valid lane, and sort all static obstacles according to their distance from the vehicle to generate an obstacle list; Determine a plurality of vertices of a first obstacle in the obstacle list, and use a vertex farthest from the vehicle among the plurality of vertices as a target vertex; Draw a perpendicular line from the target vertex to the center line of the valid lane, and determine the intersection of the perpendicular line and the center line; A search starting point is determined on the valid lane according to the intersection point.

5. A vehicle getting-out-of-trouble control device, characterized in that, include: An acquisition module, used to acquire current driving information of a vehicle and determine a driving state of the vehicle according to the current driving information; a determination module, configured to determine, when the driving state is a congested state, a plurality of lanes corresponding to the vehicle according to the current driving information; A screening module, used for screening out valid lanes from the plurality of lanes; A calculation module, used for calculating the escape position of the vehicle on the valid lane; A generating module, configured to generate an escape trajectory based on the current position in the current driving information and the escape position, and control the vehicle to travel based on the escape trajectory; Wherein, the screening module is used for: For each lane of the plurality of lanes, determining whether the lane is a target lane where the vehicle is located, and if so, determining the lane as a valid lane; if not, determining the lane as a valid lane when the lane is an adjacent lane of the target lane and the lane line of the lane is passable and determining a dynamic obstacle in the lane in front of the vehicle; The calculation module is used for: Determine a search starting point on the valid lane; obtain a search point based on the search starting point, a second preset distance and a third preset distance; and construct a virtual model of the vehicle for each search point; The virtual model and the static obstacles on the effective lane are subjected to collision detection to obtain a collision detection result corresponding to each of the search points; based on the collision detection result corresponding to each of the search points, the position of the search point corresponding to the collision detection result of no collision on the effective lane is determined as the escape position.

6. An electronic device, characterized in that, include: Memory; processor; as well as Computer programs; Wherein, the computer program is stored in the memory and is configured to be executed by the processor to implement the vehicle escape control method as described in any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, the steps of the vehicle escape control method as described in any one of claims 1 to 4 are implemented.

Images