Search

CN-117068205-B - Vehicle getting rid of poverty method, device, electronic equipment and storage medium

CN117068205BCN 117068205 BCN117068205 BCN 117068205BCN-117068205-B

Abstract

The embodiment of the disclosure discloses a vehicle getting rid of poverty method, a device, electronic equipment and a storage medium, wherein the method comprises the steps of determining a virtual terminal according to a starting point of a vehicle, a target lane, various obstacles and a preset searching distance when the vehicle is in a blocking state; the method comprises the steps of generating a sampling map based on a Frenet coordinate system according to a starting point, a virtual end point, a current lane width, other lane widths and lane lines, determining an initial path based on a fast random search tree according to the starting point, the virtual end point and the obstacles in the sampling map, determining a plurality of target getting-out points according to the starting point, the initial getting-out points and the obstacles in the initial path, and executing getting-out processing on a vehicle according to the target getting-out points. The method and the device can flexibly determine a plurality of getting rid of trapping points when the vehicle is in a blocking state, improve the rationality of the getting rid of trapping points and improve the success rate of getting rid of trapping of the vehicle.

Inventors

  • Fang Yuanyang
  • CAI QISHENG

Assignees

  • 驭势科技（北京）有限公司

Dates

Publication Date
20260512
Application Date
20230921

Claims (8)

  1. 1. A method of vehicle escape, the method comprising: Under the condition that a vehicle is in a blocking state, determining a virtual terminal according to a starting point of the vehicle, a target lane, various obstacles and a preset searching distance; Determining the length of a sampling map according to the length between the starting point and the virtual end point, determining the width of the sampling map according to the sum of the width of the current lane and the width of other lanes of each other lane corresponding to the current lane, and constructing the sampling map under a Frenet coordinate system according to the length of the sampling map and the width of the sampling map by taking the starting point as an origin; Determining an initial path based on a fast random search tree according to the starting point, the virtual end point and each obstacle in the sampling map; Determining the starting point in the initial path as a first escape point; Determining a second escaping point corresponding to the first escaping point in each initial escaping point according to the initial path and a preset trimming distance, and judging whether a connecting line of the second escaping point and the first escaping point collides with at least one obstacle; If not, removing each initial escape point located between the first escape point and the second escape point in the initial path, updating the initial path, taking the second escape point as a new first escape point, and returning to execute the step of determining a second escape point corresponding to the first escape point in each initial escape point according to the initial path and a preset trimming distance; if so, taking the initial escape point which is positioned between the second escape point and the first escape point and is adjacent to the second escape point in the initial path as a new second escape point under the condition that the second escape point and the first escape point are not adjacent in the initial path, and returning to execute the step of judging whether the connecting line of the second escape point and the first escape point collides with at least one obstacle; taking the second getting rid of the trapping point as a new first getting rid of the trapping point under the condition that the second getting rid of the trapping point is adjacent to the first getting rid of the trapping point in the initial path, and returning to the step of executing the second getting rid of the trapping point corresponding to the first getting rid of the trapping point in each initial getting rid of the trapping point according to the initial path and the preset trimming distance; When the second getting rid of the trapping point is an initial getting rid of the trapping point located at the end position in the initial path and the connection line between the second getting rid of the trapping point and the initial getting rid of the trapping point adjacent to the second getting rid of the trapping point in the initial path is not collided with each obstacle, determining the updated initial path as a target path, and determining each second getting rid of the trapping point in the target path as each target getting rid of the trapping point; Determining a target escape point adjacent to the starting point in the target path as an escape point to be optimized; projecting the to-be-optimized escaping point onto a central reference line of the target lane to obtain a first projection point; Judging whether a connecting line of the first projection point and a last target escaping point of the escaping point to be optimized or the starting point and a connecting line of the first projection point and a next target escaping point of the escaping point to be optimized collide with at least one obstacle; If not, updating the target path based on the first projection point, determining the next target escape point of the first projection point in the target path as a new escape point to be optimized, and returning to execute the step of projecting the escape point to be optimized onto a central reference line of the target lane to obtain a first projection point until the escape point to be optimized is the target escape point positioned at the end position in the target path; if so, projecting the to-be-optimized escape point onto a central reference line of a lane to which the to-be-optimized escape point belongs to obtain a second projection point, and judging whether a connection line between the second projection point and a last target escape point of the to-be-optimized escape point or the starting point and a connection line between the second projection point and a next target escape point of the to-be-optimized escape point collide with at least one obstacle; if not, updating the target path based on the second projection point, determining the next target escape point of the second projection point in the target path as a new escape point to be optimized, and returning to execute the step of projecting the escape point to be optimized onto the center reference line of the target lane to obtain a first projection point until the escape point to be optimized is the target escape point positioned at the end position in the target path; If yes, determining the next target escaping point of the escaping points to be optimized in the target path as a new escaping point to be optimized, and returning to execute the step of projecting the escaping points to be optimized onto the central reference line of the target lane to obtain a first projection point until the escaping points to be optimized are the escaping points located at the end positions in the target path; and executing the escaping treatment on the vehicle according to a plurality of target escaping points.
  2. 2. The method of claim 1, wherein the determining a virtual destination based on the starting point of the vehicle, the target lane, the obstacles, and the preset search distance comprises: Determining an initial end point on the target lane according to the initial projection point of the initial point on the central reference line of the current lane and a preset searching distance, and constructing a virtual vehicle according to the initial end point; if any obstacle does not collide with the virtual vehicle, the initial end point is taken as a virtual end point; If at least one obstacle collides with the virtual vehicle, updating the preset searching distance according to the preset expanding distance, and returning to the step of executing the initial end point on the target lane according to the starting point of the vehicle and the preset searching distance.
  3. 3. The method of claim 2, wherein said constructing a virtual vehicle from said initial endpoint comprises: Expanding the initial end point according to the size of the vehicle and a preset expansion ratio to construct a virtual vehicle; The updating the preset searching distance according to the preset expanding distance comprises the following steps: and taking the sum of the preset extension distance and the preset search distance as a new preset search distance.
  4. 4. The method according to claim 1, wherein the determining a sampling map length according to the length between the start point and the virtual end point, and determining a sampling map width according to a sum of a current lane width of the current lane and other lane widths of each other lane corresponding to the current lane, and constructing a sampling map in a Frenet coordinate system according to the sampling map length and the sampling map width with the start point as an origin, includes: Taking the current lane as an effective lane, and judging whether a lane line in the first direction of the effective lane is a broken line or not; if so, adding the adjacent lanes of the effective lane in the first direction into the effective lane, and returning to the step of executing the judgment of whether the lane line in the first direction of the effective lane is a broken line or not until the lane line in the first direction of the effective lane is a solid line; if not, judging whether the lane line in the second direction of the effective lane is a broken line or not; If so, adding the adjacent lanes of the effective lane in the second direction into the effective lane, and returning to the step of executing the judgment of whether the lane line of the effective lane in the second direction is a broken line or not until the lane line of the effective lane in the second direction is a solid line; Determining a sampling map width according to the lane width of the effective lane and a preset expansion width; determining the length of a sampling map according to the longitudinal distance between the initial projection point of the initial point on the central reference line of the current lane and the virtual end point and the preset expansion length; and taking the initial projection point as an origin of a sampling map, and establishing the sampling map under a Frenet coordinate system according to the width of the sampling map and the length of the sampling map.
  5. 5. The method of claim 1, wherein the determining an initial path based on a fast random search tree from the starting point, the virtual ending point, and each obstacle in the sampling map comprises: in the sampling map, taking the starting point as a first search point, and adding the first search point into a search tree; generating a random number, if the random number is smaller than or equal to a preset value, generating a random point in the sampling map, and if the random number is larger than the preset value, taking the virtual terminal point as the random point; If the distance between the random point and the first search point is larger than a preset step length, taking the first search point as a starting point, taking the point at the preset step length in the direction of pointing to the random point by the first search point as a new node, and if the distance between the random point and the first search point is not larger than the preset step length, taking the random point as the new node; In a Cartesian coordinate system, determining a father node corresponding to the new node according to the distance between each tree node in the search tree and the new node, and determining a collision condition according to each obstacle, a virtual vehicle corresponding to the new node and a connecting line of the new node and the father node corresponding to the new node; If the collision condition is no collision, taking the new node as a first search point, and returning to execute the step of adding the first search point into a search tree until the virtual endpoint is added into the search tree; if the collision condition is collision, returning to execute the step of generating the random number until the virtual terminal point is added into the search tree; and backtracking in the search tree according to the virtual terminal point, and determining an initial path.
  6. 6. A vehicle escape apparatus, the apparatus comprising: The virtual terminal point determining module is used for determining a virtual terminal point according to a starting point of the vehicle, a target lane, various obstacles and a preset searching distance under the condition that the vehicle is in a blocking state; The sampling map generation module is used for determining the length of a sampling map according to the length between the starting point and the virtual end point, determining the width of the sampling map according to the sum of the width of the current lane and the width of other lanes of each other lane corresponding to the current lane, and constructing the sampling map under the Frenet coordinate system according to the length of the sampling map and the width of the sampling map by taking the starting point as an origin; the initial path determining module is used for determining an initial path based on a rapid random search tree according to the starting point, the virtual end point and each obstacle in the sampling map; The target getting-off point determining module is used for determining the starting point in the initial path as a first getting-off point, determining a second getting-off point corresponding to the first getting-off point in each initial getting-off point according to the initial path and the preset trimming distance, judging whether a connecting line of the second getting-off point and the first getting-off point collides with at least one obstacle, if not, removing each initial getting-off point between the first getting-off point and the second getting-off point in the initial path, updating the second getting-off point as a new first getting-off point, returning the second getting-off point to the first getting-off point according to the initial path and the preset trimming distance, and if yes, returning the second getting-off point to the second getting-off point and the second getting-off point to the second getting-off point according to the second getting-off point in the initial path, and the second getting-off point to the second getting-off point in the initial path, and returning the second getting-off point to the first getting-off point in the initial path according to the second getting-off point in the second getting-off point when the second getting-off point is not adjacent to the first getting-off point in the initial path, determining an updated initial path as a target path and determining each second escape point in the target path as each target escape point under the condition that the second escape point is an initial escape point positioned at an end position in the initial path and a connecting line of the second escape point and an initial escape point adjacent to the second escape point in the initial path is not collided with each obstacle; The escape point optimization module is used for determining a target escape point adjacent to the starting point in the target path as an escape point to be optimized; the method comprises the steps of obtaining a first projection point by projecting a first target escape point to be optimized onto a central reference line of a target lane, judging whether a connecting line of the first projection point and a previous target escape point of the target lane or a connecting line of the first projection point and a next target escape point of the target lane collides with at least one obstacle, if not, updating the target path based on the first projection point, determining a next target escape point of the target path as a new target escape point, and returning to execute the step of projecting the first target escape point to be optimized onto the central reference line of the target lane until the first projection point is the target escape point positioned at an ending position in the target path, if not, then projecting the first target escape point onto the central reference line of the lane to which the target escape point to be optimized belongs, obtaining a second projection point, updating the target path based on the first projection point, determining whether the second projection point is the second projection point and the first target escape point, and the second projection point are the first projection point, and returning to the first projection point to the target escape point to be optimized, and the second projection point being the target escape point to be optimized, and if not being the second projection point is the target escape point positioned at the ending position, determining the next target escape point of the escape point to be optimized in the target path as a new escape point to be optimized, and returning to the step of executing the projection of the escape point to be optimized onto the center reference line of the target lane to obtain a first projection point until the escape point to be optimized is the target escape point of the target path at the end position; And the escaping execution module is used for executing escaping processing on the vehicle according to a plurality of target escaping points.
  7. 7. An electronic device, the electronic device comprising: one or more processors; a storage means for storing one or more programs; the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the vehicle getting rid of method of any of claims 1-5.
  8. 8. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when executed by a processor, implements a vehicle getting rid of poverty method according to any one of claims 1-5.

Description

Vehicle getting rid of poverty method, device, electronic equipment and storage medium Technical Field The disclosure relates to the technical field of automatic driving, and in particular relates to a vehicle getting rid of poverty method, device, electronic equipment and storage medium. Background In an automatic driving system, the vehicle gets rid of poverty to be an important part, and the purpose of getting rid of poverty is to enable the vehicle to break through some environmental constraints under the blocking state, so as to help the vehicle to continuously complete driving tasks. At present, a bias passive and conservative scheme is mainly adopted to solve the problem of getting rid of poverty, for example, a vehicle is backed up backwards and is kept a sufficient distance from a front obstacle in a blocking state and then detours, or a getting rid of poverty point is generated based on scene rules, and a planning method in free space is adopted to help the vehicle to move to the getting rid of poverty point so as to restore to a normal state. The above escape mode can only deal with simpler conditions, has poorer flexibility, has the condition that the generated escape point exceeds the road boundary, and seriously affects the safety and the running efficiency of the vehicle. Disclosure of Invention In order to solve the above technical problems or at least partially solve the above technical problems, the embodiments of the present disclosure provide a vehicle getting rid of poverty method, device, electronic device and storage medium, which can flexibly determine a plurality of getting rid of poverty points when a vehicle is in a blocking state, improve the rationality of the getting rid of poverty points, and improve the success rate of getting rid of poverty of the vehicle. In a first aspect, embodiments of the present disclosure provide a vehicle escape method, the method comprising: Under the condition that a vehicle is in a blocking state, determining a virtual terminal according to a starting point of the vehicle, a target lane, various obstacles and a preset searching distance; generating a sampling map based on a Frenet coordinate system according to the starting point, the virtual terminal point, the current lane width, other lane widths and each lane line; Determining an initial path based on a fast random search tree according to the starting point, the virtual end point and each obstacle in the sampling map; determining a plurality of target escape points according to the starting point in the initial path, each initial escape point in the initial path and each obstacle; and executing the escaping treatment on the vehicle according to a plurality of target escaping points. In a second aspect, embodiments of the present disclosure also provide a vehicle escape apparatus, the apparatus comprising: The virtual terminal point determining module is used for determining a virtual terminal point according to a starting point of the vehicle, a target lane, various obstacles and a preset searching distance under the condition that the vehicle is in a blocking state; The sampling map generation module is used for generating a sampling map based on a Frenet coordinate system according to the starting point, the virtual end point, the current lane width, other lane widths and each lane line; the initial path determining module is used for determining an initial path based on a rapid random search tree according to the starting point, the virtual end point and each obstacle in the sampling map; The target escape point determining module is used for determining a plurality of target escape points according to the starting point in the initial path, each initial escape point in the initial path and each obstacle; And the escaping execution module is used for executing escaping processing on the vehicle according to a plurality of target escaping points. In a third aspect, the disclosed embodiments also provide an electronic device comprising one or more processors, a storage device for storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the vehicle getting rid of poverty method as described above. In a fourth aspect, the disclosed embodiments also provide a computer readable storage medium having stored thereon a computer program which when executed by a processor implements a vehicle getting rid of poverty method as described above. According to the vehicle getting-out method, when a vehicle is in a blocking state, a virtual end point is determined according to a starting point, a target lane, barriers and preset searching distances of the vehicle, so that the vehicle is convenient to search and plan getting-out points, a sampling map is generated based on a Frenet coordinate system according to the starting point, the virtual end point, the current lane width, other lane widths and lane lines, the situation that the vehic