Object position estimating system, object position estimating apparatus, object position estimating method, and object position estimating program

• US 8,111,876 B2
• Filed: 07/21/2010
• Issued: 02/07/2012
• Est. Priority Date: 11/05/2008
• Status: Active Grant

• Current Assignee: Panasonic Corporation (Panasonic Holdings Corporation)
• Original Assignee: Panasonic Corporation (Panasonic Holdings Corporation)
• Inventors: Tanigawa, Toru
• Primary Examiner(s): Mariam, Daniel
• Assistant Examiner(s): Bitar, Nancy
• Application Number:
  US12/840,593
• Publication Number:
  US 20100284571A1
• Time in Patent Office: 566 Days
• Field of Search: 382/102, 382/105, 382/106, 707/5, 702/95, 702/150
• US Class Current: 382/106
• CPC Class Codes: G01B 11/03 by measuring coordinates of points
• CPC Class Codes: G01B 21/00 Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
• CPC Class Codes: G06T 7/74 involving reference images or patches
• CPC Class Codes: G06T 7/75 involving models
• CPC Class Codes: G06T 7/77 using statistical methods

Abstract

A plurality of pieces of observation raw data observed by an observation unit are subjected to an object identifying process, and by using a parameter determined by a parameter determination unit in accordance with a period of time between a point of object identifying process completion time of one piece of data and process completion scheduled time of another piece of data, a position estimating process of the object is carried out from the data that has been subjected to the object identifying process, and the position of the object relating to the object ID is estimated based upon the object ID and positional candidates acquired by an object identifying unit.

Claims

1. An object position estimating system comprising:

an observation unit configured to observe a state of an environment that includes an object placed therein to acquire first and second observation raw data;

an object identifying configured to acquire an ID of the object and positional candidates of the object based upon object identifying data for each of IDs of objects, preliminarily stored in an object database, and the observation raw data of the respective first and second observation raw data, and to respectively store the acquired ID of the object and information of the positional candidates of the object in an observation database;

an object identifying process completion time predicting configured to acquire a number of the objects located within the environment based upon the respective observation raw data of the first and second observation raw data, and to predict a completion scheduled time of an object identifying process, based upon a time at which each number of the objects is acquired, the number of objects respectively acquired, and a period of time required for the object identifying process in which each of IDs of the object and the positional candidates of the object per object, preliminarily stored, are acquired, and to store the completion scheduled time of each of the object identifying processes in the observation database;

an object position estimating configured to acquire the ID of the object and the positional candidates of the object stored in the observation database, and to estimate a position of the object based upon the acquired ID of the object, the acquired positional candidates of the object, and a dispersion state of the positional candidates of the object; and

a parameter determination configured to refer to the completion scheduled time of the object identifying process of the second observation raw data stored in the observation database, at a time when the object identifying unit has acquired the ID of the object and the positional candidates of the object after completion of the object identifying process of the first observation raw data, and to determine as to whether or not the completion scheduled time of the object identifying process for the second observation raw data is stored in the observation database based upon the time that has been referred to, as well as, upon determination of a fact that the completion scheduled time of the object identifying process of the second observation raw data has been stored therein, to determine, by reference to a period of time required for carrying out a position estimating process for estimating the position of the object for each of the objects, preliminarily stored, a parameter by which a position estimating process of the object is carried out on the first observation raw data that has been subjected to the object identifying process, during a period of time from a time of the reference to the completion scheduled time of the object identifying process of the second observation raw data, while, upon determination of a fact that the completion scheduled time of the object identifying process of the second observation raw data has not been stored therein, to set a predetermined reference parameter as the parameter by which the position estimating process is carried out on the first observation raw data that has been subjected to the object identifying process,

wherein by using the parameter determined by the parameter determination unit, the object position estimating unit carries out the position estimating process on the object contained in the data that has been subjected to the object identifying process, from the data, and estimates the position of the object relating to the object ID based upon the object ID and the positional candidates acquired by the object identifying unit.

2. The object position estimating system as claimed in claim 1, wherein the object identifying data is shape information of the object, and the observation raw data is image data photographed by using a camera, or range data acquired by a range-measuring sensor, or positional data acquired by a tag reader.

3. The object position estimating system as claimed in claim 2, wherein the data obtained by the camera or the range-measuring sensor is subjected to an object identifying process through a template matching process;

the object identifying unit stores a matching score outputted in the template matching process in the observation database; and

the parameter determination unit determines an object to be subjected to the position estimating process in an order of a higher matching score.

4. The object position estimating system as claimed in claim 1, wherein upon using a particle filter, the object position estimating unit preliminarily prepares relational information between a processing period of time per object and a number of particles, and calculates a number of objects that are processable by using the particle filter based upon a period of time from the time of the reference to the completion scheduled time of the object identifying process of the second observation raw data, and in a case where the calculated number of objects is same as the number of the objects identified by the object identifying process, the particle filter, as it is, is used for the process,

while in a case where the calculated number of objects is smaller than the number of the objects identified by the object identifying process, the process is carried out with the number of particles per object being increased, and

in a case where the calculated number of objects is greater than the number of the objects identified by the object identifying process, the process is carried out with the number of particles per object being reduced.

5. The object position estimating system as claimed in claim 1, further comprising:

a display configured to display a result of estimation by the object position estimating unit,

wherein referring to the observation database, the display unit displays a completion scheduled time of a next object identifying process as an output time of result of next estimation of the object position estimating unit, while the display unit displays a time at which observation raw data relating to an output of current estimation result of the object position estimating unit as a point of time at which the current estimation result of the object position estimating unit is obtained.

6. An object position estimating system comprising:

a data input unit capable of inputting data relating to at least an ID and a position of an object that has been subjected to an object identifying process and completion scheduled time of the object identifying process that is inputted after completion of the object identifying process of the data;

an observation database to store the data relating to the ID and position of the object received from the data input unit and the data relating to the input scheduled time;

an object position estimating configured to estimate the position of the object based upon the data relating to the ID and position of the object stored in the observation database; and

a parameter determination configured to determine whether or not by reference to the completion scheduled time of the object identifying process stored in the observation database, the completion scheduled time of the object identifying process to be completed at an earliest time from a time of the reference is stored, and upon determination that the completion scheduled time of the object identifying process of the second observation raw data has been stored, by reference to a period of time required for carrying out the position estimating process for estimating the position of the object per object preliminarily stored, to determine a parameter by which the position estimating process of the object is carried out during a period of time from a time of the reference to the completion scheduled time of the object identifying process of the second observation raw data, while, upon determination of a fact that the completion scheduled time of the object identifying process of the second observation raw data has not been stored therein, to set a predetermined reference parameter as the parameter by which the position estimating process is carried out on the object,

wherein by using the parameter determined by the parameter determination unit, the object position estimating unit carries out a position estimating process on the object based upon the data relating to the ID and position of the object stored in the observation database so as to estimate the position of the object.

7. An object position estimating method comprising:

observing a state of an environment so that a plurality of pieces of observation raw data are acquired by an observation unit;

based upon the plurality of pieces of observation raw data, calculating an ID and a position of an object located within an observation range of the observation unit by using an object identifying unit so as to carry out the object identifying process;

predicting a point of time at which the object identifying process to be carried out by the object identifying unit is completed based upon each of the plurality of pieces of the observation raw data by using an object identifying a process completion time predicting unit so as to store the resulting times in an observation database;

referring to the completion scheduled time of the object identifying process stored in the observation database, determining as to whether or not the completion scheduled time of the object identifying process to be completed at an earliest point of time from the time of the reference is stored therein, by using a parameter determination unit;

upon determination of a fact that the completion scheduled time of the object identifying process of the second observation raw data has been stored therein, determining, by reference to a period of time required for carrying out the position estimating process for estimating the position of the object per object, preliminarily stored, a parameter by which the position estimating process is carried out on the object during a period of time from the time of the reference to the completion scheduled time of the object identifying process of the second observation raw data,

while, upon determination of a fact that the completion scheduled time of the object identifying process of the second observation raw data has not been stored therein, setting a predetermined reference parameter as the parameter by which the position estimating process is carried out on the object, by using the parameter determination unit,

wherein by using the parameter determined by the parameter determination unit, the object position estimating unit estimates the position of the object from the data relating to the ID and the position of the object calculated in the object identifying process.

8. A non-transitory computer readable medium storing an object position estimating program allowing a computer to carry out functions of:

based upon a plurality of pieces of observation raw data acquired by observing a state of an environment by using an observation unit, calculating an ID and a position of an object located within an observation range of the observation unit so as to carry out an object identifying process;

predicting a point of time at which the object identifying process to be carried out by the object identifying function is completed based upon each of the plurality of pieces of the observation raw data so as to store the resulting times in an observation database;

referring to the completion scheduled time of the object identifying process stored in the observation database, determining as to whether or not the completion scheduled time of the object identifying process to be completed at an earliest point of time from a time of the reference is stored therein, and upon determination of a fact that the completion scheduled time of the object identifying process of the second observation raw data has been stored therein, determining, by reference to a period of time required for carrying out the position estimating process for estimating the position of the object per object, preliminarily stored, a parameter by which the position estimating process is carried out on the object during a period of time from a time of the reference to the completion scheduled time of the object identifying process of the second observation raw data, while upon determination of a fact that the completion scheduled time of the object identifying process of the second observation raw data has not been stored therein, setting a predetermined reference parameter as the parameter by which the position estimating process is carried out on the object, wherein by using the parameter determined by the parameter determination unit, carrying out an object position estimating function for estimating the position of the object, from the data relating to the ID and the position of the object calculated in the object identifying process by the object identifying process function.