Methods, apparatus, and medium for estimating pose of mobile robot using particle filter

• US 8,055,382 B2
• Filed: 09/14/2007
• Issued: 11/08/2011
• Est. Priority Date: 11/16/2006
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Myeong, Hyeon, Bang, Seok-won, Lee, Su-jinn, Kong, Dong-geon
• Primary Examiner(s): Nguyen; John
• Assistant Examiner(s): FEI, JORDAN S
• Application Number:
  US11/898,779
• Publication Number:
  US 20080119961A1
• Time in Patent Office: 1,516 Days
• Field of Search: 700/245, 700/253, 700/258, 700/259, 700/250, 701/23, 701/25, 701/26, 701/207, 701/208, 901/1, 901/46, 901/47
• US Class Current: 700/253
• 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: G05D 1/0272 comprising means for registering the travel distance, e.g. revolutions of wheels measuring distance traversed on the ground by vehicles, e.g. using odometers G01C22/00
• CPC Class Codes: G05D 1/0274 using mapping information stored in a memory device navigation using map-matching G01C21/30

Abstract

Disclosed is a method, apparatus, and medium for estimating a pose of a moving robot using a particle filter. A method for estimating a pose of a moving robot using a particle filter according to an embodiment of the invention includes a detecting a change in pose of the mobile robot and calculating a pose of the current particle by applying the detected change in pose to the previous particle, predicting the probability of the pose of the current particle and obtaining a weight of the current particle on the basis of range data obtained by a sensor and map information, resampling the current particle on the basis of the weight, and adjusting the weight in consideration of an error of the sensor

Claims

1. A method for estimating a pose of a mobile robot using a particle filter, the method comprising:

sampling current particle on basis of initial position of the mobile robot;

detecting a change in pose of the mobile robot and calculating pose of the current particle by applying the detected change in pose to the previous particle;

predicting the probability of the pose of the current particle and obtaining a weight of the current particle on the basis of range data obtained by a sensor and map information;

resampling the current particle on the basis of the weight; and

adjusting the weight in consideration of an error of the sensor,

wherein the detecting of the change in pose of the mobile robot and the calculating of the pose of the current particle by applying the detected change in pose to the previous particle comprises updating the pose of the current particle in a direction where the position probability increases by using a hill climbing method on the basis of a relative coordinate system according to a heading angle of the mobile robot, and

wherein the hill climbing method causes a step size to be changed adaptively to an error covariance ellipse of the odometer model that detects the change in pose such that the step size increases in a direction where the robot proceeds.

2. The method of claim 1, wherein the detecting of the change in pose of the Mobile robot and the calculating of the pose of the current particle by applying the detected change in pose to the previous particle comprises updating the pose of the current particle in a direction where the position probability increases by using a hill climbing method.

3. The method of claim 1, wherein the pose includes coordinate values in a two-dimensional coordinate system on the plane and a heading angle.

4. The method of claim 1, wherein the sensor comprises a structured light sensor.

5. The method of claim 1, wherein the resampling removes a particle having the low weight and performs a sampling operation on the basis of a particle having the high weight.

6. The method of claim 1, wherein the change in pose is detected by an encoder or a gyroscope.

7. A method for estimating a pose of a mobile robot using a particle filter, the method comprising:

sampling current particle on basis of initial position of the mobile robot;

detecting a change in pose of the mobile robot and calculating pose of the current particle by applying the detected change in pose to the previous particle;

predicting the probability of the pose of the current particle and obtaining a weight of the current particle on the basis of range data obtained by a sensor and map information;

resampling the current particle on the basis of the weight; and

adjusting the weight in consideration of an error of the sensor,

wherein the adjusting of the weight in consideration of the error by the sensor comprises adjusting the weight according to an error characteristic value that is defined on the basis of the distance on the map at the position of the current particle and the distance measured by the sensor.

8. The method of claim 7, wherein the detecting of the change in pose of the mobile robot and the calculating of the pose of the current particle by applying the detected change in pose to the previous particle comprises updating the pose of the current particle in a direction where the position probability increases by using a hill climbing method.

9. The method of claim 7, wherein the change in pose is detected by an encoder or a gyroscope.

10. The method of claim 7,wherein the pose includes coordinate values in a two-dimensional coordinate system on the plane and a heading angle.

11. The method of claim 7, wherein the sensor comprises a structured light sensor.

12. The method of claim 7, wherein the resampling removes a particle having the low weight and performs a sampling operation on the basis of a particle having the high weight.