Method and system for maneuvering a mobile robot

• US 5,525,882 A
• Filed: 10/24/1994
• Issued: 06/11/1996
• Est. Priority Date: 10/25/1993
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Ishikawa, Shigeki, Hazeki, Shinichiro, Echigo, Tomio, Asaka, Shunichi
• Primary Examiner(s): Martin, David S.
• Application Number:
  US08/328,030
• Time in Patent Office: 596 Days
• Field of Search: 318/568.12, 318/568.11, 318/568.16, 318/587, 901/1, 901/46, 901/47, 364/424
• US Class Current: 318/568.16
• 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: H04N 13/10 Processing, recording or transmission of stereoscopic or multi-view image signals
• CPC Class Codes: H04N 13/239 using two 2D image sensors having a relative position equal to or related to the interocular distance H04N13/243 takes precedence
• CPC Class Codes: H04N 2013/0081 Depth or disparity estimation from stereoscopic image signals

Abstract

A method and system for maneuvering a mobile robot are disclosed, which include a mobile robot having a plurality of sensors, including a sensor which detects the presence of an object and a sensor which detects image information related to the object. Utilizing data acquired by the plurality of sensors, elements of the object are determined. A hypothesis expressing the configuration of the object is then generated from the elements interpreted from the data. The hypothesis is generated by referencing a table in which interpretations of data acquired by the plurality of sensors are associated with possible configurations of objects in response to the generated hypothesis, the mobile robot is moved along a path which avoids the object.

Claims

1. A control system for maneuvering a mobile robot comprising:

a memory;

a plurality of sensors, including at least a first sensor for detecting proximity of objects and a second sensor for detecting image information related to objects, wherein each of said plurality of sensors generates sensor data in response to a detection of an object;

means for interpreting sensor data generated by said plurality of sensors to determine a plurality of interpretations of said detected object, wherein each of said plurality of interpretations is determined from sensor data generated by a single one of said plurality of sensors and indicates whether a surface component of said detected object was sensed by that sensor;

a table stored within said memory, wherein said table associates interpretations of sensor data generated by said plurality of sensors with predetermined configurations of objects;

means for deriving a model of said detected object from said plurality of interpretations by referencing said table; and

means, responsive to said derivation of said model of said detected object, for moving said mobile robot.

2. The control system for maneuvering a mobile robot of claim 1, wherein said means for deriving a model classifies configurations of objects by surface features characteristic of said objects, and wherein said table stores, for each of said surface features, combinations of interpretations of sensor data generated by each of said plurality of sensors.

3. The control system for maneuvering a mobile robot of claim 1, and further comprising:

means for verifying said model by comparing said model to interpretations of sensor data generated subsequent to said derivation of said model by at least one of said plurality of sensors; and

means for discarding portions of said model to obtain an updated model, wherein said discarded portions of said model contradict said interpretations of sensor data generated subsequent to said derivation of said model by said at least one of said plurality of sensors.

4. The control system for maneuvering a mobile robot of claim 3, and further comprising:

means for thereafter deriving a second model of said particular object from said updated model and sensor data generated by at least one of said plurality of sensors.

5. A system for determining configurations of objects within an environment of a mobile robot, comprising:

a memory;

a plurality of sensors, including at least a first sensor for detecting proximity of objects and a second sensor for detecting image information related to objects, wherein each of said plurality of sensors generates sensor data in response to a detection of an object;

means for interpreting sensor data generated by said plurality of sensors to determine a plurality of interpretations of said detected object, wherein each of said plurality of interpretations is determined from sensor data generated by a single one of said plurality of sensors and indicates whether a surface component of said detected object was sensed by that sensor;

a table stored within said memory, wherein said table associates interpretations of sensor data generated by said plurality of sensors with predetermined configurations of objects; and

means for deriving a model of said detected object from said plurality of interpretations by referencing said table.

6. The system for determining configurations of objects within an environment of a mobile robot of claim 5, wherein said means for deriving a model classifies configurations of objects by surface features characteristic of said objects, and wherein said table stores, for each of said surface features, combinations of interpretations of sensor data generated by each of said plurality of sensors.

7. The system for determining configurations of objects within an environment of a mobile robot of claim 5, and further comprising:

means for verifying said model by comparing said model to interpretations of sensor data generated subsequent to said derivation of said model by at least one of said plurality of sensors; and

means for discarding portions of said model to obtain an updated model, wherein said discarded portions of said model contradict said interpretations of sensor data generated subsequent to said derivation of said model by said at least one of said plurality of sensors.

8. The system for determining configurations of objects within an environment of a mobile robot of claim 7, and further comprising:

means for thereafter deriving a second model of said particular object utilizing said updated model and sensor data generated by at least one of said plurality of sensors.

10. The method for maneuvering a mobile robot of claim 6, and further comprising the steps of:

verifying said model by comparing said model to interpretations of sensor data generated by at least one of said plurality of sensors subsequent to said derivation of said model; and

discarding portions of said model to obtain an updated first model, wherein discarded portions of said model contradict said interpretations of sensor data generated by at least one of said plurality of sensors subsequent to said derivation of said model.

9. A method for maneuvering a mobile robot, said mobile robot having a plurality of sensors including a first sensor for detecting proximity of objects and a second sensor for detecting image information related to objects, said method comprising:

detecting an object proximate to said mobile robot utilizing said first sensor;

detecting image information related to said detected object utilizing said second sensor;

interpreting sensor data generated by said plurality of sensors to determine a plurality of interpretations of said detected object, wherein each of said plurality of interpretations is determined from sensor data generated by a single one of said plurality of sensors;

deriving a model of said detected object from said plurality of interpretations, wherein said model is derived by referencing a table in which combinations of interpretations of sensor data generated by said first and said second sensors are associated with predetermined configurations of objects; and

in response to derivation of said model, moving said mobile robot.

11. The method for maneuvering a mobile robot of claim 9, and further comprising:

thereafter, deriving a second model of said particular object utilizing said updated model and sensor data generated by at least one of said plurality of sensors.

12. A method for determining an environment of a mobile robot, said mobile robot having a plurality of sensors, including a first sensor for detecting proximity of objects and a second sensor for detecting image information related to objects, said method comprising:

detecting an object proximate to said mobile robot utilizing said first sensor;

detecting image information related to said detected object utilizing said second sensor;

interpreting sensor data generated by said first and said second sensors to determine a plurality of interpretations of said detected object, wherein each of said plurality of interpretations is determined from data generated by a single one of said plurality of sensors; and

deriving a model of said detected object from said plurality of interpretations, wherein said model is derived by referencing a table in which combinations of interpretations of sensor data generated by said first and said second sensors are associated with predetermined configurations of objects.

13. The method for determining an environment of a mobile robot of claim 12, and further comprising the steps of:

verifying said model by comparing said model to interpretations of sensor data generated subsequent to said derivation of said model by at least one of said plurality of sensors; and

discarding portions of said model to obtain an updated model, wherein discarded portions of said model contradict said interpretations of sensor data generated subsequent to said derivation of said model by at least one of said plurality of sensors.

14. The method for determining an environment of a mobile robot of claim 13, and further comprising:

thereafter, deriving a second model of said detected object utilizing said updated model and sensor data generated by at least one of said plurality of sensors.