Methods and systems for mobile-agent navigation

• US 9,625,908 B2
• Filed: 10/12/2015
• Issued: 04/18/2017
• Est. Priority Date: 09/03/2014
• Status: Active Grant

• Current Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.)
• Original Assignee: Sharp Laboratories of America Incorporated (Hon Hai Precision Industry Co., Ltd.)
• Inventors: Liao, Miao, Xu, Xinyu, van Beek, Petrus J. L.
• Primary Examiner(s): Dager, Jonathan M
• Application Number:
  US14/881,005
• Publication Number:
  US 20160062359A1
• Time in Patent Office: 554 Days
• Field of Search: None
• CPC Class Codes: G01C 21/20 Instruments for performing navigational calculations G01C21/24, G01C21/26 take precedence
• CPC Class Codes: G01C 21/34 Route searching; Route guidance
• CPC Class Codes: G05D 1/0088 characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours using knowledge based models G06N5/00
• CPC Class Codes: G05D 1/0234 using optical markers or beacons optical beacons per se G01S1/70
• CPC Class Codes: G05D 1/0246 using a video camera in combination with image processing means
• CPC Class Codes: G05D 1/0253 extracting relative motion information from a plurality of images taken successively, e.g. visual odometry, optical flow determining position or orientation from images G06T7/70

Abstract

Aspects of the present invention are related to methods and systems for autonomous navigation using visual landmark recognition.

Claims

1. A method for autonomous navigation, said method comprising:

capturing an image;

performing landmark detection on said captured image, thereby identifying one or more detected landmark objects;

computing a distance and a bearing for each detected landmark object in said one or more detected landmark objects;

selecting a nearest landmark object, from said one or more detected landmark objects, using said computed distances;

computing a mobile-agent speed control value and a mobile-agent steering control value for said nearest landmark object using said computed distance for said nearest landmark object and said computed bearing for said nearest landmark object; and

sending said mobile-agent speed control value and said mobile-agent steering control value to a control component of a mobile agent.

2. A method as described in claim 1, wherein said performing landmark detection comprises:

receiving a captured image;

generating a plurality of candidate landmark regions in said captured image; and

determining an association for each landmark region, in said plurality of candidate landmark regions, with a landmark object in a plurality of existing landmark objects.

3. A method as described in claim 2, wherein said performing landmark detection further comprises indicating any unobserved landmark objects from said plurality of existing landmark objects.

4. A method as described in claim 3, wherein said performing landmark detection further comprises deleting a dematerialized landmark object from said plurality of existing landmark objects, wherein said dematerialized landmark object is a landmark object from said plurality of landmark objects indicated as unobserved for a number of frames.

5. A method as described in claim 2, wherein said performing landmark detection further comprises:

creating a first new landmark object associated with a first candidate landmark region, in said plurality of candidate landmark regions, when said determining an association for each landmark region does not generate an association for said first candidate landmark region; and

adding said first new landmark object to said plurality of existing landmark objects.

6. A method as described in claim 2, wherein said generating a plurality of candidate landmark regions comprises:

generating a plurality of initial candidate landmark regions based on a color model;

performing a merging operation on said plurality of initial candidate landmark regions to form a plurality of merged candidate landmark regions; and

performing a filtering operation on said plurality of merged candidate landmark regions to form said plurality of candidate landmark regions.

7. A method as described in claim 2 further comprising updating a feature associated with a first existing landmark object with a feature value computed for a first candidate landmark region, wherein said first candidate landmark region is associated with said first existing landmark object.

8. A method as described in claim 1, wherein said mobile-agent speed control value is:

proportional to a normal driving speed for said mobile agent when said computed distance for said nearest landmark object is between a first distance threshold and a second distance threshold;

equal to zero when said computed distance for said nearest landmark object is less than or equal to said first distance threshold; and

equal to said normal driving speed when said computed distance for said nearest landmark object is greater than said second threshold.

9. A method as described in claim 1, wherein said mobile-agent steering control value is proportional to said computed bearing for said nearest landmark object.

10. A mobile agent comprising:

a camera for capturing an image;

a control component; and

a non-transitory computer-readable medium encoded with a computer program code for implementing a method for autonomous navigation, said method comprising;

performing landmark detection on said captured image, thereby identifying one or more detected landmark objects;

computing a distance and a bearing for each detected landmark object in said one or more detected landmark objects;

selecting a nearest landmark object, from said one or more detected landmark objects, using said computed distances;

computing a mobile-age speed control value and a mobile-agent steering control value for said nearest landmark object using said computed distance for said nearest landmark object and said computed bearing for said nearest landmark object; and

sending said mobile-agent speed control value and said mobile-agent steering control value to said control component.