Computer-implemented method and system for locating an indoor object
First Claim
1. A computer-implemented method for locating an indoor object based on a radio frequency (RF) position signal from an RF position estimation system, and at least one detecting signal from a motion sensor array upon detection of movement of the indoor object, the RF position signal carrying information of an RF position on a position plane estimating a position of the indoor object, said method comprising:
- a) receiving the RF position signal and the detecting signal;
b) providing information of a located region on the position plane with reference to the RF position and the detecting signal;
c) obtaining a motion sensor position on the position plane according to the information of the located region;
d) adopting a covariance intersection (CI) estimator for offline training and generating weights for the RF position and the motion sensor position;
e) applying the weights to the RF position and the motion sensor position, respectively; and
f) combining the weighted RF position and the weighted motion sensor position to generate a fused position on the position plane corresponding to the position of the indoor object;
wherein when only one detecting signal is received in step a), step b) is performed by providing information of a predetermined sensing area of the motion sensor array on the position plane in which the RF position falls, followed by defining the predetermined sensing area as the located region; and
wherein when more than one detecting signals are received in step a), step b) includes the sub-steps of;
b-1) providing information of a plurality of pre-determined sensing areas of the motion sensor array on the position plane, such that at least two of the pre-determined sensing areas overlap each other so as to divide the pre-determined sensing areas into an overlapped region and non-overlapped regions; and
b-2) determining one of the non-overlapped regions and the overlapped region, in which the RF position falls, as the located region.
1 Assignment
0 Petitions
Accused Products
Abstract
A computer-implemented method for locating an indoor object includes: a) receiving an RF coordinate signal and a detecting signal; b) providing information of a located region with reference to the RF coordinate and the detecting signal; c) obtaining a motion sensor coordinate according to the information of the located region; d) providing weights for the RF coordinate and the motion sensor coordinate; e) applying the weights to the RF coordinate and the motion sensor coordinate; and f) combining the weighted RF coordinate and the weighted motion sensor coordinate to generate a fused coordinate corresponding to the position of the indoor object. A system for locating an indoor object is also disclosed.
9 Citations
14 Claims
-
1. A computer-implemented method for locating an indoor object based on a radio frequency (RF) position signal from an RF position estimation system, and at least one detecting signal from a motion sensor array upon detection of movement of the indoor object, the RF position signal carrying information of an RF position on a position plane estimating a position of the indoor object, said method comprising:
-
a) receiving the RF position signal and the detecting signal; b) providing information of a located region on the position plane with reference to the RF position and the detecting signal; c) obtaining a motion sensor position on the position plane according to the information of the located region; d) adopting a covariance intersection (CI) estimator for offline training and generating weights for the RF position and the motion sensor position; e) applying the weights to the RF position and the motion sensor position, respectively; and f) combining the weighted RF position and the weighted motion sensor position to generate a fused position on the position plane corresponding to the position of the indoor object; wherein when only one detecting signal is received in step a), step b) is performed by providing information of a predetermined sensing area of the motion sensor array on the position plane in which the RF position falls, followed by defining the predetermined sensing area as the located region; and wherein when more than one detecting signals are received in step a), step b) includes the sub-steps of; b-1) providing information of a plurality of pre-determined sensing areas of the motion sensor array on the position plane, such that at least two of the pre-determined sensing areas overlap each other so as to divide the pre-determined sensing areas into an overlapped region and non-overlapped regions; and b-2) determining one of the non-overlapped regions and the overlapped region, in which the RF position falls, as the located region. - View Dependent Claims (2, 3, 4, 5, 6, 7)
-
-
8. A system for locating an indoor object, comprising:
-
a radio frequency (RF) position estimation system including at least three reference node RF transceivers, a blind node RF transceiver, and a localization module, wherein each of said reference node RF transceivers correspond to a pre-determined reference node coordinate on a position plane and are placed at a pre-determined position, said blind node RF transceiver being carried by an indoor object, said localization module being configured to receive Received Signal Strength Index (RSSI) values related to distances between said at least three reference node RF transceivers and said blind node RF transceiver, and to generate a RF position on the position plane according to the pre-determined reference node positions on the position plane and the RSSI values; a motion sensor position estimation system including a motion sensor array and a localization module, said motion sensor array including at least one motion sensor configured to output a detecting signal upon detection of movement of the indoor object, said localization module of the motion sensor position estimation system being configured to receive the RF position signal and the detecting signal, to provide information of a located region on the position plane with reference to the RF position and the detecting signal, and to obtain a motion sensor position on the position plane according to the information of the located region; and a fusion system including a weight generation module and a fused position calculation module, said weight generation module of said fusion system adopting a covariance intersection (CI) estimator for offline training and generating the weights for the RF position and the motion sensor position, said fused position calculation module being configured to receive the RF position and the motion sensor position, to apply the weights to the RF position and the motion sensor position, respectively, and to combine the weighted RF position and the weighted motion sensor position to generate a fused position on the position plane corresponding to the position of the indoor object. - View Dependent Claims (9, 10, 11, 12, 13, 14)
-
Specification