Search

EP-4737861-A2 - METHOD AND APPARATUS, CONTROLLER, AND COMPUTER PROGRAM PRODUCT FOR LANE FUSION

EP4737861A2EP 4737861 A2EP4737861 A2EP 4737861A2EP-4737861-A2

Abstract

Examples of the present disclosure relate to a method and an apparatus, a controller, and a computer program product for lane fusion. The method includes determining a map boundary of at least one lane corresponding to a target road section based on map data. The method further comprises obtaining a sensing boundary for the at least one lane based on sensing data from at least one sensor. The method further comprises identifying an abnormal boundary of the target road section based on a match of the map boundary to the sensing boundary. In this way, not only can the matching between map data with different data structures or representations and the sensing data be efficiently processed, but also can the matching results be used to identify the abnormal boundaries that may be present in the map data and the sensing data, thereby promoting performance improvements in driving functionality and facilitating more rational and safe driving decisions for vehicles.

Inventors

  • WANG, Leichen
  • LI, Xinrun

Assignees

  • Robert Bosch GmbH

Dates

Publication Date
20260506
Application Date
20251029

Claims (15)

  1. A method (200) for lane fusion, comprising: determining (210) a map boundary of at least one lane corresponding to a target road section based on map data; obtaining (220) a sensing boundary for the at least one lane based on sensing data from at least one sensor; and identifying (230) an abnormal boundary of the target road section based on a match of the map boundary to the sensing boundary.
  2. The method (200) according to claim 1, wherein the map boundary is indicated by an out-of-bounds boundary of a map boundary on one side of the at least one lane corresponding to the target road section, the method comprising: searching for a target road section corresponding to the sensing boundary in the map data, the map data including a road section, a lane, and a boundary; the out-of-bounds boundary acquired by at least one of the following: in response to a distance between a first boundary and a second boundary in a vehicle travel direction of the searched target road section being less than a first distance threshold, forming the out-of-bounds boundary by splicing the first boundary with the second boundary; in response to a total length of the searched target road section being greater than a second distance threshold, forming the out-of-bounds boundary by cropping a portion of the searched target road section that exceeds the second distance threshold.
  3. The method (200) according to claim 2, wherein the target road section corresponding to the sensing boundary comprises at least one of: a first road section overlapping the sensing boundary; or a predetermined number of second road sections upstream and downstream of the first road section.
  4. The method (200) according to claim 1, wherein the match of the map boundary to the sensing boundary comprises: determining a similarity between an out-of-bounds boundary of the map data and the sensing boundary of the sensing data; and constructing a boundary topology indicating a matching relationship between the out-of-bounds boundary and the sensing boundary based on the calculated similarity.
  5. The method (200) according to claim 4, wherein determining the similarity comprises: determining a point distance between a sample point on the out-of-bounds boundary and a nearest point on the sensing boundary; and obtaining a boundary distance between the out-of-bounds boundary and the sensing boundary by weighting the determined point distance.
  6. The method (200) according to claim 4, wherein the boundary topology is constructed by: creating a start node and an end node; creating an out-of-bounds node corresponding to each out-of-bounds boundary, and a sensing boundary node corresponding to each sensing boundary; and matching the out-of-bounds node having a minimum node distance and the sensing boundary node based on a node distance indicating a boundary distance between the out-of-bounds boundary and the sensing boundary.
  7. The method (200) according to claim 6, further comprising: in response to a boundary distance between a first out-of-bounds boundary and a first sensing boundary that are unmatched being greater than a third distance threshold corresponding to a lane width, pre-matching the first out-of-bounds boundary with the first sensing boundary as an abnormal match.
  8. The method (200) according to claim 7, further comprising: obtaining a match result between the out-of-bounds boundary and the sensing boundary by filtering out an abnormal match between the out-of-bounds boundary and the sensing boundary.
  9. The method (200) according to claim 8, wherein filtering out the abnormal match comprises: in response to a second out-of-bounds boundary matching with a plurality of second sensing boundaries, retaining a match between one second sensing boundary having a minimum boundary distance to the second out-of-bounds boundary; and in response to a boundary distance between a third out-of-bounds boundary and a third sensing boundary that have been matched being greater than the third distance threshold, deleting a pre-match between the third out-of-bounds boundary and the third sensing boundary.
  10. The method (200) according to claim 1, wherein the sensing data comprises first sensing data from a first sensor and second sensing data from a second sensor, the method (200) further comprising: performing a first fusion based on a first match of the map data with the first sensing data; performing a second fusion based on a second match of the map data with the second sensing data; and performing an adjustment on fusion results of the first fusion and the second fusion, respectively.
  11. The method (200) according to claim 10, further comprising: assigning a confidence score for the map data, the first sensing data, or the second sensing data based on the first match and the second match.
  12. The method (200) according to claim 11, wherein the abnormal boundary includes a false negative boundary, and identifying the abnormal boundary comprises: in response to a fourth sensing boundary from the first sensing data and a fifth sensing boundary associated with the fourth sensing boundary from the second sensing data not matching any out-of-bounds boundaries, assigning a high confidence score to a presence of a false negative boundary in the map data at locations of the fourth sensing boundary and the fifth sensing boundary; and in response to a fourth out-of-bounds boundary matching only with a sixth sensing boundary from the first sensing data or the second sensing data, assigning a high confidence score to a presence of a false negative boundary in unmatched sensing data at locations of the fourth out-of-bounds boundary and the sixth sensing boundary.
  13. An apparatus (1000) for lane fusion, comprising: a map boundary determination module (1010) configured to determine a map boundary of at least one lane corresponding to a target road section based on map data; a sensing boundary acquisition module (1020) configured to acquire a sensing boundary for the at least one lane based on the sensing data from at least one sensor; and a boundary matching module (1030) configured to identify an abnormal boundary of the target road section based on a match of the map boundary to the sensing boundary.
  14. A controller (1100), comprising: at least one processor (1101); and a memory (1102) coupled to the at least one processor (1101) and having instructions stored thereon that, when executed by the at least one processor (1101), cause the controller (1100) to perform the method according to any one of claims 1 to 11.
  15. A computer program product, the computer program product being tangibly stored on a computer-readable medium (1102) and comprising computer-executable instructions that, when executed by a processor (1101) of a computer, cause the computer to perform the method according to any one of claims 1 to 12.

Description

Technical Field Embodiments of the present disclosure relate generally to the field of driving, and in particular, to a method and an apparatus, a controller, and a computer program product for lane fusion. Background Accurate modeling of complex traffic scenes is not only a key step in achieving accurate planning and effective control of driving systems (such as autonomous driving (AD systems) and advanced driving assistance systems (ADAS)), but also an important foundation for improving road safety and reducing the risk of traffic accidents, etc. In order to keep the vehicle moving in the correct position, a lane detection function is provided, with an intention to identify and track markings in the road to achieve accurate navigation and safe driving of the vehicle. There are a variety of detection and sensing methods, such as camera-based lane detection, and laser radar (LiDAR)-based lane detection. Summary of the Invention Embodiments of the present disclosure provide a method and an apparatus, a controller, and a computer program product for lane fusion. A first aspect of the present disclosure provides a method for lane fusion. The method includes determining a map boundary of at least one lane corresponding to a target road section based on map data. The method further includes obtaining a sensing boundary for the at least one lane based on sensing data from at least one sensor. The method further includes identifying an abnormal boundary of the target road section based on a match of the map boundary to the sensing boundary. A second aspect of the present disclosure provides an apparatus for lane fusion. The apparatus includes a map boundary determination module configured to determine a map boundary of at least one lane corresponding to a target road section based on map data. The apparatus further includes a sensing boundary acquisition module, which is configured to acquire a sensing boundary for the at least one lane based on sensing data from at least one sensor. The apparatus further includes a boundary matching module configured to identify an abnormal boundary of the target road section based on a match of the map boundary to the sensing boundary. A third aspect of the present disclosure provides a controller. The controller includes at least one processor. The controller further includes a memory coupled to the at least one processor and having instructions stored thereon that, when executed by the at least one processor, cause a device to execute the steps of the method according to the first aspect of the present disclosure. A fourth aspect of the present disclosure provides a vehicle. The vehicle includes the controller according to the third aspect of the present disclosure. A fifth aspect of the present disclosure provides a computer program product, which is tangibly stored on a computer-readable medium and includes computer-executable instructions that, when executed by a processor of a computer, cause the computer to execute the steps of the method according to the first aspect of the present disclosure. A sixth aspect of the present disclosure provides a machine-readable storage medium having instructions stored thereon that, when executed by a processor, cause the machine to execute the steps of the method according to the first aspect of the disclosure. Brief Description of the Drawings The above-described and other purposes, features, and advantages of the present disclosure will become clearer by more detailed description of the exemplary embodiments of the present disclosure in conjunction with the accompanying drawings. In illustrative embodiments of the present disclosure, the same or similar reference numerals generally represent the same or similar parts, components, etc. FIG. 1 illustrates a schematic diagram of an exemplary environment in which the method and/or device according to embodiments of the present disclosure may be implemented.FIG. 2 illustrates a flow chart of a method for lane fusion according to embodiments of the present disclosure;FIG. 3 illustrates a diagram of a lane fusion process according to embodiments of the present disclosure;FIG. 4 illustrates a diagram of an example data representation utilizing out-of-bounds boundary map data and sensing data according to embodiments of the present disclosure;FIG. 5 illustrates a diagram of a process for out-of-bounds boundary acquisition according to embodiments of the present disclosure;FIG. 6 illustrates a schematic diagram of a boundary topology process according to embodiments of the present disclosure;FIG. 7 illustrates a schematic diagram of a constructed boundary topology according to embodiments of the present disclosure;FIG. 8 illustrates a schematic diagram of a multi-modal fusion process according to embodiments of the present disclosure;FIG. 9 illustrates a schematic diagram of an abnormal boundary identification process based on multi-modal sensing data according to embodiments of the present dis