Self-navigating vehicle equipped with lane boundary recognition system

• US 5,517,412 A
• Filed: 09/16/1994
• Issued: 05/14/1996
• Est. Priority Date: 09/17/1993
• Status: Expired due to Term

• Current Assignee: Honda Giken Kogyo Kabushiki Kaisha (Honda Motor Company)
• Original Assignee: Honda Giken Kogyo Kabushiki Kaisha (Honda Motor Company)
• Inventors: Unoura, Kiyozumi
• Primary Examiner(s): Chin, Gary
• Application Number:
  US08/307,367
• Time in Patent Office: 606 Days
• Field of Search: 364/424.02, 364/460, 364/461, 364/449, 180/167, 180/168, 180/169, 318/587, 348/113, 348/116, 348/118, 348/135, 348/139, 340/435, 340/903
• US Class Current: 701/23
• CPC Class Codes: G05D 1/0246 using a video camera in combination with image processing means
• CPC Class Codes: G05D 1/0251 extracting 3D information from a plurality of images taken from different locations, e.g. stereo vision stereoscopic image analysis H04N13/00; depth recovery from images G06T7/593
• CPC Class Codes: G05D 1/0255 using acoustic signals, e.g. ultra-sonic singals sonar systems designed for anti-collision purposes G01S15/93
• CPC Class Codes: G06V 10/457 by analysing connectivity, e.g. edge linking, connected component analysis or slices
• CPC Class Codes: G06V 20/588 Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road

Abstract

A lane boundary recognition system for a self-navigating vehicle which enables the vehicle to self-navigate on an expressway or other such roadway having lane boundaries while simultaneously identifying and discriminating the continuity of the lane boundaries. Vehicle position to the lane boundaries are determined at times T0 and T1. A vehicle locus occurred in a period between the times is estimated and the position at the time T0 is corrected. A difference between said corrected position and said position at the time T1 is calculated and is adding to lane boundary numbers and by comparing them with each other, it is determining if the lane boundaries recognized at the times T0 and T1 are continuous. If determined to be continuous, the same number is assigned and is stored in a common memory.

Claims

1. A system for recognizing a lane boundary for use in a self navigating vehicle, comprising:

first means for capturing a view of a lane of a roadway on which said vehicle travels to obtain an image of said lane;

second means for extracting at least one line segment from said image;

third means for recognizing said lane at a first time and a second time based upon said at least one line segment;

fourth means for estimating a locus of said vehicle during a period between said first and second times;

fifth means for determining whether or not said lane recognized at said first time and said lane recognized at said second time are continuous and for assigning a same lane identifier when said lane recognized at said first time and said lane recognized at said second time are determined to be continuous; and

sixth means for storing information regarding said lane identifier.

2. A system according to claim 1, wherein said fifth means includes:

seventh means for determining a first value indicative of a position of said vehicle with respect to said lane recognized at said first time and a second value indicative of a position of said vehicle based upon said locus of said vehicle; and

wherein

said fifth means compares said first and second values and, based on a result of that comparison, determines whether or not said lane recognized at said first time and said lane recognized at said second time are continuous.

3. A system according to claim 2, wherein

said third means determines a boundary of said lane based on said at least one line segment to recognize said lane at said first time and at said second time and assigns a lane number to said lane boundary;

said seventh means determines said first and second values with respect to said lane boundary having an assigned number; and

said fifth means obtains a difference between said first and second values and adds said difference to said assigned number to increment said assigned number and, by comparing a magnitude of said assigned number at said first time to a magnitude of said assigned number at said second time determine whether or not said lane recognized at said first time and said lane recognized at said second time are continuous.

4. A system according to claim 1, wherein said roadway has a plurality of lanes.

5. A system according to claim 1, wherein said vehicle is equipped with a steering control system which enables said vehicle to self-navigate along said lane, and said steering control system controls steering of said vehicle based on said stored information.

6. A system for recognizing a lane boundary for a self-navigating vehicle, comprising:

first means for capturing a view of a plurality of lanes of a roadway on which said vehicle travels to obtain an image of said plurality of lanes;

second means for extracting a plurality of line segments from said image to recognize a plurality of boundaries of said lanes at a first time T0 and a second time T1;

third means for assigning a reference number to one of said lane boundaries;

fourth means for determining a distance between said vehicle and each of said lane boundaries and dividing each said distance by a value corresponding to a width of each of said lanes to produce and assign a quotient value to each of said lane boundaries;

fifth means for determining a position of said vehicle relative to said lane boundary assigned said reference number at said first time T0 and said second time T1;

sixth means for estimating a locus of said vehicle during a period between said first time T0 and said second time T1 to provide a corrected position of said vehicle at said first time T0 and for obtaining a difference between said corrected position at said first time T0 and said position of said vehicle at said second time T1;

seventh means for adding said difference to said quotient value of each of said lane boundaries to produce an incremented quotient value for each of said lane boundaries and, by comparing said incremented quotient values with each other, to determine whether or not said lane boundaries recognized at said first time T0 and said lane boundaries recognized at said second time T1 are continuous, and assigning a same lane identifier when said lane boundaries recognized at said first time T0 and said lane boundaries recognized at said second time T1 are determined to be continuous; and

eighth means for storing information concerning said lane identifier.

7. A system according to claim 6, wherein said vehicle is equipped with a steering control system which enables said vehicle to self-navigate along one of said plurality of lanes, and said steering control system controls steering of said vehicle based on said stored information.

8. A lane boundary recognition system for use in a vehicle navigation system, said boundary recognition system comprising:

a camera for generating an image of a roadway;

an image processing processor coupled to said camera for converting said image generated by said camera into a digital data set representing said image and storing said data set in a memory; and

an image evaluation processor coupled to said memory, said image evaluation processor being programmed to;

extract at a first time T0 and a second time T1 a plurality of line segments each representing a lane boundary from said image of said roadway;

assign a lane boundary number to each of said line segments;

determine if any of said line segments extracted at said first time T0 and said line segments extracted at said second time T1 represent a continuous lane; and

assign a same lane identifier to any of said line segments representing said continuous lane.

9. A method for use in an image evaluation processor of a vehicle self-navigation system, said method comprising the steps of:

extracting at a first time T0 and a second time T1 a plurality of line segments each representing a lane boundary from an image of a roadway;

assigning a lane boundary number to each of said line segments;

determining if any of said line segments extracted at said first time T0 and said line segments extracted at said second time T1 represent a continuous lane; and

assigning a same lane identifier to any of said line segments representing said continuous lane.