Claims:

A processor-implemented method for target localization, the method comprising:
enabling movement of an Unmanned Aerial Vehicle (UAV) along a trajectory of a sampling pattern in a Region of Interest (ROI) comprising a target, via one or more hardware processors, the target being stationary, and the trajectory comprises circular trajectory of radius R;
collecting, during the movement of the UAV along the trajectory, a plurality of received signal strength (RSS) samples corresponding to RF signals emitted by the target, via the one or more hardware processors, the plurality of RSS samples taken at equal angular separation f along the circular trajectory and corresponding to a plurality of target locations in the ROI;
performing localization by using a Support Vector Regression (SVR) model, based on the plurality of RSS samples to obtain an initial estimate of target location, via the one or more hardware processors, the SVR model previously trained using a plurality of RSS training samples of the RF signals emitted by the target;
guiding movement of the UAV towards the target location obtained by the initial estimate, via the one or more hardware processors; and
repeating the localization at the target location obtained by the initial estimate to obtain a subsequent estimate of the target location using the SVR model, via the one or more hardware processors,
wherein the subsequent estimate of the target location is relatively precise than the initial estimate of the target location, comprising:

The processor implemented method of claim 1, wherein training the SVR model comprises:
enabling movement of the UAV along the trajectory of the sampling pattern in the ROI, wherein the center of the sampling pattern is fixed at the center of the ROI and is taken as origin;
collecting the plurality of RSS training samples (si) corresponding to a plurality of known target locations (Bj) during the movement of the UAV along the trajectory to obtain a training dataset for the SVR model, the trajectory during the training being a circular trajectory of radius R, and the plurality of RSS training samples taken at equal angular separation f and corresponding to the plurality of known target locations in the ROI;
providing the plurality of RSS training samples (si) to the SVR model for determining SVR model parameters; and
determining, by the SVR model, optimal SVR model parameters (a, a*) for x co-ordinate and y co-ordinate as below:
¦("max" -1/2 ?_(i,j=0)¦?(a_xi-a_xi^* )(a_xj-a_xj^* )?(s_i )^T ?("s" _j ) ?-??_(i=1)^l¦?(a_xi-a_xi^* )+?_(i=1)^l¦x_i (a_xi-a_xi^* ) ?)

¦("max" -1/2 ?_(i,j=0)¦?(a_yi-a_yi^* )(a_yj-a_yj^* ) ??(s_i )?^T ?("s" _j)?-??_(i=1)^l¦?(a_yi-a_yi^* )+?_(i=1)^l¦y_yi (a_yi-a_yi^* ) ?)

The processor implemented method of claim 2, wherein the SVR model is selected from amongst a first SVR model and a second SVR model based on a threshold value of magnitude of the RSS samples, wherein the threshold value is determined during the training and depends on a distance (r) of the UAV with respect to the target location.

The processor implemented method of claim 1, wherein guiding the movement of the UAV towards the target location obtained by the initial estimate comprises:
heading a velocity vector of the UAV towards the target location obtained by the initial estimate by making the heading angle of the UAV equal to the line of sight (LOS) angle between the UAV and the target location, wherein a lateral acceleration for the UAV is given by the equation:
a-?_M=(-K(a_M-?_LOS ))/V_M +?-?_LOS "\,"
where, aM ?LOS are the heading angle and LOS angle respectively, and K and is a guidance gain.

The processor implemented method of claim 1, wherein obtaining the subsequent estimate of the target location comprises:
enabling the UAV to traverse along the trajectory with the radius R and collecting the plurality of RSS samples at equal regular separation;
arranging the plurality of RSS samples with respect to a reference line and storing in a RSS strength vector;
computing an angle ?’ made by a first RSS sample of the plurality of RSS samples with a reference line at a center of the trajectory;
providing the RSS strength vector to the trained SVR model to estimate the initial estimated location of the target based on the equations:
x_i=?_(i=1)^l¦?(a_ix-a_ix^* )k("s" _i,"s" _j )+b_x ?
y_i=?_(i=1)^l¦?(a_iy-a_iy^* )k("s" _iy,"s" _j )+b_y ?

wherein, xi and yi represents estimated relative location of the target with respect to a center of a localization circle rotated by an angle –(?’-?); and
rotating the estimated relative location of the target based on the equation:
¦(x-() ^ y-() ^ )=[¦(cos?(?^'-?)&-sin?(?^'-?)@sin?(?^'-?)&cos?(?^'-?) )][¦(x-() ^ '@y-() ^ ')]"\,"
wherein,( x-() ^ ^',y-() ^ ') represents the coordinates of the target obtained from the SVR model, and (x-() ^ ,y-() ^ ) represents the initial estimated location of the target after rotation.

The processor implemented method of claim 5, further comprising updating the subsequent estimate of the target location for each subsequent RSS sample, wherein updating the subsequent estimate comprises:
providing a set of subsequent RSS samples to the SVR model; and
obtaining the subsequent estimate of the target location based on the set of subsequent RSS samples.

A system (901) for target localization, comprising:
one or more memories (915); and
one or more hardware processors (902), the one or more memories coupled to the one or more hardware processors, wherein the one or more hardware processors are capable of executing programmed instructions stored in the one or more memories to:
enable movement of an Unmanned Aerial Vehicle (UAV) along a trajectory of a sampling pattern in a Region of Interest (ROI) comprising a target, the target being stationary, and the trajectory comprises circular trajectory of radius R;
collect, during the movement of the UAV along the trajectory, a plurality of received signal strength (RSS) samples corresponding to RF signals emitted by the target, the plurality of RSS samples taken at equal angular separation f along the circular trajectory and corresponding to a plurality of target locations in the ROI;
perform localization by using a Support Vector Regression (SVR) model, based on the plurality of RSS samples to obtain an initial estimate of target location, the SVR model previously trained using a plurality of RSS training samples of the RF signals emitted by the target;
guide movement of the UAV towards the target location obtained by the initial estimate; and
repeat the localization at the target location obtained by the initial estimate to obtain a subsequent estimate of the target location using the SVR model,
wherein the subsequent estimate of the target location is relatively precise than the initial estimate of the target location.

The system of claim 7, wherein training the SVR model comprises:
enable movement of the UAV along the trajectory of the sampling pattern in the ROI, wherein the center of the sampling pattern is fixed at the center of the ROI and is taken as origin;
collect the plurality of RSS training samples (si) corresponding to a plurality of known target locations (Bj) during the movement of the UAV along the trajectory to obtain a training dataset for the SVR model, the trajectory during the training being a circular trajectory of radius R, and the plurality of RSS training samples taken at equal angular separation f and corresponding to the plurality of known target locations in the ROI;
provide the plurality of RSS training samples (si) to the SVR model for determining SVR model parameters; and
determine, by the SVR model, optimal SVR model parameters (a, a*) for x co-ordinate and y co-ordinate as below:
¦("max" -1/2 ?_(i,j=0)¦?(a_xi-a_xi^* )(a_xj-a_xj^* )?(s_i )^T ?("s" _j ) ?-??_(i=1)^l¦?(a_xi-a_xi^* )+?_(i=1)^l¦x_i (a_xi-a_xi^* ) ?)

¦("max" -1/2 ?_(i,j=0)¦?(a_yi-a_yi^* )(a_yj-a_yj^* ) ??(s_i )?^T ?("s" _j)?-??_(i=1)^l¦?(a_yi-a_yi^* )+?_(i=1)^l¦y_yi (a_yi-a_yi^* ) ?)

The system of claim 8, wherein the SVR model is selected from amongst a first SVR model and a second SVR model based on a threshold value of magnitude of the RSS samples, wherein the threshold value is determined during the training and depends on a distance (r) of the UAV with respect to the target location.

The system of claim 7, wherein guiding the movement of the UAV towards the target location obtained by the initial estimate comprises:
heading a velocity vector of the UAV towards the target location obtained by the initial estimate by making the heading angle of the UAV equal to the line of sight (LOS) angle between the UAV and the target location, wherein a lateral acceleration for the UAV is given by the equation:
a-?_M=(-K(a_M-?_LOS ))/V_M +?-?_LOS "\,"
where, aM ?LOS are the heading angle and LOS angle respectively, and K and is a guidance gain.

The system of claim 7, wherein obtaining the subsequent estimate of the target location comprises:
enabling the UAV to traverse along the trajectory with the radius R and collecting the plurality of RSS samples at equal regular separation;
arranging the plurality of RSS samples with respect to a reference line and storing in a RSS strength vector;
computing an angle ?’ made by a first RSS sample of the plurality of RSS samples with a reference line at a center of the trajectory;
providing the RSS strength vector to the trained SVR model to estimate the initial estimated location of the target based on the equations:
x_i=?_(i=1)^l¦?(a_ix-a_ix^* )k("s" _i,"s" _j )+b_x ?
y_i=?_(i=1)^l¦?(a_iy-a_iy^* )k("s" _iy,"s" _j )+b_y ?

wherein, xi and yi represents estimated relative location of the target with respect to a center of a localization circle rotated by an angle –(?’-?); and
rotating the estimated relative location of the target based on the equation:
¦(x-() ^ y-() ^ )=[¦(cos?(?^'-?)&-sin?(?^'-?)@sin?(?^'-?)&cos?(?^'-?) )][¦(x-() ^ '@y-() ^ ')]"\,"
wherein,( x-() ^ ^',y-() ^ ') represents the coordinates of the target obtained from the SVR model, and (x-() ^ ,y-() ^ ) represents the initial estimated location of the target after rotation.

The system of claim 7, further comprising updating the subsequent estimate of the target location for each subsequent RSS sample, wherein updating the subsequent estimate comprises:
providing a set of subsequent RSS samples to the SVR model; and
obtaining the subsequent estimate of the target location based on the set of subsequent RSS samples.
, Description:FORM 2

THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENT RULES, 2003

COMPLETE SPECIFICATION
(See Section 10 and Rule 13)

Title of invention:

SYSTEM AND METHOD FOR TARGET LOCALIZATION

Applicant

Tata Consultancy Services Limited
A company Incorporated in India under the Companies Act, 1956
Having address:
Nirmal Building, 9th floor,
Nariman point, Mumbai 400021,
Maharashtra, India

The following specification particularly describes the invention and the manner in which it is to be performed.


TECHNICAL FIELD
The disclosure herein generally relates to target localization, and, more particularly, to system and method for target localization.

BACKGROUND
Location awareness refers to a technology that facilitates provisioning of information about physical location of a device (for example, a target) to another device or application. Location awareness is a critical requirement for many location-aware applications such as surveillance, search and rescue, wireless sensor networks (WSN), and so on.
Various conventional techniques have been proposed to tackle location awareness depending on the applications at hand. Although Global Positioning System (GPS) has been around for many years, its usage is restricted by the availability of satellite signals thus making it unusable in indoor and urban area localization where the GPS signal is severely affected by intervening structure. In addition, GPS based localization requires use of special hardware and higher power requirements due to which it cannot be used in applications demanding low cost and low power requirements as usually is the requirement in the case of WSNs.

SUMMARY
Embodiments of the present disclosure present technological improvements as solutions to one or more of the above-mentioned technical problems recognized by the inventors in conventional systems. For example, in one embodiment, a processor implemented method for target localization is provided. The method includes enabling movement of an Unmanned Aerial Vehicle (UAV) along a trajectory of a sampling pattern in a Region of Interest (ROI) comprising a target, via one or more hardware processors. Herein, the target is stationary, and the trajectory comprises circular trajectory of radius R. Further, the method includes collecting, during the movement of the UAV along the trajectory, a plurality of received signal strength (RSS) samples corresponding to RF signals emitted by the target, via the one or more hardware processors, the plurality of RSS samples taken at equal angular separation f along the circular trajectory and corresponding to a plurality of target locations in the ROI. Furthermore, the method includes performing localization by using a Support Vector Regression (SVR) model, based on the plurality of RSS samples to obtain an initial estimate of target location, via the one or more hardware processors, the SVR model previously trained using a plurality of RSS training samples of the RF signals emitted by the target. Moreover, the method includes guiding movement of the UAV towards the target location obtained by the initial estimate, via the one or more hardware processors. Also, the method includes repeating the localization at the target location obtained by the initial estimate to obtain a subsequent estimate of the target location using the SVR model, via the one or more hardware processors. Herein, the subsequent estimate of the target location is relatively precise than the initial estimate of the target location.
In another aspect, a system for target localization is provided. The system includes one or more memories; and one or more hardware processors, the one or more memories coupled to the one or more hardware processors, wherein the one or more hardware processors are capable of executing programmed instructions stored in the one or more memories to enable movement of an Unmanned Aerial Vehicle (UAV) along a trajectory of a sampling pattern in a Region of Interest (ROI) comprising a target, the target being stationary, and the trajectory comprises circular trajectory of radius R. Further, the one or more hardware processors are capable of executing programmed instructions stored in the one or more memories to collect, during the movement of the UAV along the trajectory, a plurality of received signal strength (RSS) samples corresponding to RF signals emitted by the target, the plurality of RSS samples taken at equal angular separation f along the circular trajectory and corresponding to a plurality of target locations in the ROI. Further, the one or more hardware processors are capable of executing programmed instructions stored in the one or more memories to perform localization by using a Support Vector Regression (SVR) model, based on the plurality of RSS samples to obtain an initial estimate of target location, the SVR model previously trained using a plurality of RSS training samples of the RF signals emitted by the target. Furthermore, the one or more hardware processors are capable of executing programmed instructions stored in the one or more memories to guide movement of the UAV towards the target location obtained by the initial estimate. Moreover, the one or more hardware processors are capable of executing programmed instructions stored in the one or more memories to repeat the localization at the target location obtained by the initial estimate to obtain a subsequent estimate of the target location using the SVR model. Herein, the subsequent estimate of the target location is relatively precise than the initial estimate of the target location.
In yet another aspect, a non-transitory computer-readable medium having embodied thereon a computer program for executing a method for target localization is provided. The method includes enabling movement of an Unmanned Aerial Vehicle (UAV) along a trajectory of a sampling pattern in a Region of Interest (ROI) comprising a target, via one or more hardware processors. Herein, the target is stationary, and the trajectory comprises circular trajectory of radius R. Further, the method includes collecting, during the movement of the UAV along the trajectory, a plurality of received signal strength (RSS) samples corresponding to RF signals emitted by the target, via the one or more hardware processors, the plurality of RSS samples taken at equal angular separation f along the circular trajectory and corresponding to a plurality of target locations in the ROI. Furthermore, the method includes performing localization by using a Support Vector Regression (SVR) model, based on the plurality of RSS samples to obtain an initial estimate of target location, via the one or more hardware processors, the SVR model previously trained using a plurality of RSS training samples of the RF signals emitted by the target. Moreover, the method includes guiding movement of the UAV towards the target location obtained by the initial estimate, via the one or more hardware processors. Also, the method includes repeating the localization at the target location obtained by the initial estimate to obtain a subsequent estimate of the target location using the SVR model, via the one or more hardware processors. Herein, the subsequent estimate of the target location is relatively precise than the initial estimate of the target location.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed principles.
FIG. 1A illustrates a networking environment implementing a system for target localization according to some embodiments of the present disclosure.
FIG. 1B illustrates a discovery stage and a pursuit stage followed by a UAV for target localization according to some embodiments of the present disclosure.
FIG. 2 is a flow diagram illustrating a method for target localization in accordance with some embodiments of the present disclosure.
FIG. 3 illustrates example geometry of a UAV and a target for target localization according to some embodiments of the present disclosure.
FIG. 4 illustrates sampling pattern and target location used for training a SVR model for target localization, in accordance with some embodiments of the present disclosure.
FIG. 5 illustrates sampling pattern and target location used for training a SVR model for target localization, in accordance with some embodiments of the present disclosure.
FIG. 6 illustrates effect of rotated sampling pattern for target localization, in accordance with some embodiments of the present disclosure.
FIG. 7 illustrates trajectory of a UAV along with sampling locations and estimated target locations for target localization in accordance with some embodiments of the present disclosure.
FIG. 8 illustrates a variation of target location estimation error for target localization in accordance with some embodiments of the present disclosure.
FIG. 9 illustrates a block diagram of a system for target localization in accordance with some embodiments of the present disclosure.

DETAILED DESCRIPTION OF EMBODIMENTS
Location awareness is a critical requirement for many location-aware applications such as surveillance, search and rescue, WSN, and so on. The use of GPS technology for location awareness is restricted by the availability of satellite signals thus making it unusable in indoor and urban area localization where the GPS signal is severely affected by intervening structure. In addition, GPS based localization requires use of special hardware and higher power requirements due to which it cannot be used in applications demanding low cost and low power requirements as usually is the requirement in the case of WSNs. To overcome the aforementioned issues, other conventional localization systems have been proposed based on different techniques. Said proposed techniques can be broadly classified as range-dependent or range-free approaches.
Range-dependent schemes are those in which location is computed by estimating the absolute pairwise distance between two nodes, using relative location information such as Received Signal Strength (RSS), Time Difference of Arrival (TDoA), Time of Arrival (ToA), Time of Flight (ToF) or Angle of Arrival (AoA). These range-dependent methods rely on signal propagation model (in case of RSS) which are sensitive to several environmental parameters, or may require very expensive hardware such as accurate clocks or array of antennas (TDoA, ToA, ToF, and AoA), and the cost incurred in doing so usually does not justify the application. Range free systems overcome these disadvantages by using signal strength information to directly estimate location of a sensor and/or target (or target) using machine learning or optimization based algorithms.
RSS based localization has been explored extensively in different applications, especially where sensors with limited computational capabilities are to be used. Fingerprinting method is one of the early approaches that was developed for localization. It consists of offline phase in which the RSS measurements from the access points (AP) at different reference points in the operation area are computed and stored along with its coordinates in the database. In the localization phase the sensor measures the RSS from the APs and looks for similar pattern in the database and chooses the closest match and returns the approximate location. This method demands higher storage capacity and computational power. Kernel based techniques for localization using RSS has been proposed in literature due to their ability to better capture the nonlinear relationships between the RSS and the location. These methods also comprise of the training phase and the localization phase, with difference being in training phase, where RSS measurements collected are used to develop regression model which is later used during the localization phase. SVR based method has been proposed for indoor localization using Wi-fi signals. Another conventional method has used SVR method for localization using global system of mobile communications (GSM). Ensemble implementation of SVR is proposed to enhance the performance of target localization in terms of robustness to noise. It achieves this by dividing the dataset into subsets and by building sub-predictors for each subsets and then combining their estimates with proper weights to get the final estimates.
In another conventional technique, RSS based localization with unknown transmit power and unknown path loss exponent (PLE) is considered. In In yet another conventional technique for target localization, differential RSS information is used for localization. In still another conventional technique, estimation of the sensor location is obtained solving localization as one complex least square kernel regression problem, where the sensors 2D location is represented as a complex number. Additionally, one conventional technique addresses the localization in wireless sensor networks (WSN) using signal strengths with kernel spectral regression to estimate the location. WSN localization, characterized by fully automated parameter selection for easy adaptability to different scenario without user intervention is proposed in another conventional technique. Localization based on kernel based regression techniques that exist in the literature consider access points (or mobile towers) that are fixed at a single location in the region of interest (ROI). Thus, the model developed during the training phase always gives the target location with respect to the fixed AP’s.
Various embodiments described herein provide method and system for target localization in a cost efficient and power efficient manner. In an embodiment, the system utilizes support vector regression (SVR) technique to localize a target by using the received signal strength (RSS) measurements of the emitted RF signal. The system utilizes a circular localization pattern to collect the RSS samples and utilize the same to train a SVR model for estimating the target location directly. In the aforementioned description, the target is assumed to be a beacon; hence the terms beacon and target are used interchangeably. It will however be understood that the disclosed embodiments can be used for localization of any device that is capable of emitting RF signals and is not limited to a beacon. Any such device may be referred to as a ‘target’ in context of various embodiments disclosed herein.
Exemplary embodiments are described with reference to the accompanying drawings. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. Wherever convenient, the same reference numbers are used throughout the drawings to refer to the same or like parts. While examples and features of disclosed principles are described herein, modifications, adaptations, and other implementations are possible without departing from the spirit and scope of the disclosed embodiments. It is intended that the following detailed description be considered as exemplary only, with the true scope and spirit being indicated by the following claims.
Referring now to the drawings, and more particularly to FIG. 1A through 9, where similar reference characters denote corresponding features consistently throughout the figures, there are shown preferred embodiments and these embodiments are described in the context of the following exemplary system and/or method.
Referring to FIG. 1A, a network implementation 100 of a system 102 for target localization is illustrated, in accordance with an example embodiment. The system 102 is configured to localize the location of a target (for example a target) which is source of RF signals using a single unmanned aerial vehicle (UAV). For instance, in case of an oil spill in a geographical area, the UAV may be utilized to determine areas of concentration of oil in a region of interest (ROI) within the geographical area by using the UAV. In such a scenario, various targets or beacons may be spread across the ROI, which measure the presence of oil and transmits the data to the UAV over RF signals. The received signal strength of (RSS) corresponding to the RF signals emitted by the target may be captured by the system 102 to localize the target/beacon (targets).
In an embodiment, the system 102 utilizes a Support Vector Regression (SVR) technique to directly localize the target by using the RSS of the RF signal as the input. The system 102 exploits the ability of the SVR to learn the non-linear relationship between the RSS and the target location by transforming it into the higher dimensional feature space, where the relationships between them becomes linear.
In an embodiment, the system 102 utilizes a sampling trajectory following a circular pattern for collecting the RSS measurements of the emitted RF signal. At each sampling location in the circular pattern, an average of p-continuous samples is used as input to reduce the effect of noise and a Kalman filter is used to get a more precise location of the sensor during localization.
In an embodiment, the system 102 may operate in a discovery and pursuit stages. The discovery and pursuit stages for target localization are described further with reference to FIG. 1B. As illustrated in FIG. 1B, during the discovery stage, the system 102 obtains an initial estimate of the target location by using a trained SVR model. In the discovery stage the system 102 enables/commands the UAV to follow a circular sampling pattern to obtain the initial estimate of the location of the target. For the brevity of description, the ‘initial estimate of the location’ of the target may be referred to as ‘initial location’ of the target. In an embodiment, the initial location of the target refers to a relative location of the target with respect to the sampling pattern, as will be described with reference to FIG. 3.
Upon estimation of the initial location, in the pursuit stage, the UAV is made to traverse towards the target location identified during the discovery stage, namely the initial location. The system repeats estimation of the initial location of the target based on the RSS samples of signal emitted by the target, and estimates a subsequent location of the target such that the subsequent location is relatively precise than the initial location of the target.
Although the present subject matter is explained considering that the system 102 is implemented on a server, it may be understood that the system 102 may also be implemented in a variety of computing systems, such as a laptop computer, a desktop computer, a notebook, a workstation, a mainframe computer, a server, a network server, a cloud-based computing environment and the like. It will be understood that the system 102 may be accessed by multiple users through one or more user devices 104-1, 104-2…104-N, collectively referred to as user devices 104 hereinafter, or applications residing on the user devices 104. In one implementation, the system 102 may include a cloud-based computing environment in which a user may operate individual computing systems configured to execute remotely located applications. Examples of the user devices 104 may include, but are not limited to, a portable computer, a personal digital assistant, a handheld device, and a workstation. The user devices 104 are communicatively coupled to the system 102 through a network 108.
The servers, such as the server 106, include but are not limited to application servers, database servers, computation farms, data centers, virtual machines, cloud computing devices, mail or web servers and the like. The server 106 includes one or more computing devices or machines capable of operating one or more Web-based and/or non-Web-based applications that may be accessed by other computing devices (e.g. client devices, other servers) via the network 108. One or more servers 106 may be front end Web servers, application servers, and/or database servers.
The server 106 may include a cluster of a plurality of servers which are managed by a network traffic device such as a firewall, load balancer, web accelerator, gateway device, router, hub and the like. In an aspect, the server 106 may implement a version of Microsoft® IIS servers, RADIUS servers and/or Apache® servers, although other types of servers may be used and other types of applications may be available on the servers 106.
In an embodiment, the network implementation 100 includes one or more databases such as a repository 110, communicatively coupled to the servers 106. The repository 110 may be configured to allow storage and access to data, files or otherwise information utilized or produced by the system 102. Herein, it is assumed that the repository 110 is embodied in computing devices configured external to the servers 106. It will however be noted that in alternative embodiments, the repository 110 may be embodied in the servers 106.
In one implementation, the network 108 may be a wireless network, a wired network or a combination thereof. The network 108 can be implemented as one of the different types of networks, such as intranet, local area network (LAN), wide area network (WAN), the internet, and the like. The network 108 may either be a dedicated network or a shared network. The shared network represents an association of the different types of networks that use a variety of protocols, for example, Hypertext Transfer Protocol (HTTP), Transmission Control Protocol/Internet Protocol (TCP/IP), Wireless Application Protocol (WAP), and the like, to communicate with one another. Further the network 108 may include a variety of network devices, including routers, bridges, servers, computing devices, storage devices, and the like.
Referring now to FIG. 2 an example flow-diagram of a method 200 for target localization, according to some embodiments of present disclosure. The method 200 may be described in the general context of computer executable instructions. Generally, computer executable instructions can include routines, programs, hands, components, data structures, procedures, modules, functions, etc., that perform particular functions or implement particular abstract data types. The method 200 may also be practiced in a distributed computing environment where functions are performed by remote processing devices that are linked through a communication network. The order in which the method 200 is described is not intended to be construed as a limitation, and any number of the described method blocks can be combined in any order to implement the method 200, or an alternative method. Furthermore, the method 200 can be implemented in any suitable hardware, software, firmware, or combination thereof. In an embodiment, the method 200 depicted in the flow chart may be executed by a system, for example, the system 102 of FIG. 1. In an example embodiment, the system 102 may be embodied in an exemplary computer system, for example computer system 901 (FIG. 9). The method 200 of FIG. 2 will be explained in more detail below by taking reference from FIGS. 2-9.
Referring to FIG. 2, the method 200 initiates at 202, where movement of a UAV along a trajectory of a sampling pattern in a ROI is enabled. The ROI includes a target which may be a stationary target. In an embodiment, the UAV is enabled to move around a circular trajectory. Herein, the UAV collects RSS samples of the RF signals emitted by the target to perform localization of the target location by obtaining an initial estimate of the target location. An example of relative locations of the UAV and the target is described further with reference to FIG. 3
Referring now to FIG. 3, the geometry 300 of a UAV 302 and a target 304 are illustrated in accordance with an example embodiment. The UAV’s task is to localize the target 304 by using only the available signal strength s and its known location M. RSS obtained by the UAV 302 is denoted by s. M denotes the location of the UAV 302 and it is assumed to be known, B represents the location of the target 304. Mi represents the location of UAV 302 when ith RSS sample is collected and the corresponding RSS sample is denoted by si and the vector storing the RSS measurements is represented by s. In an embodiment, the geometry of the UAV 302 and the target 304 is planar. Said geometry is extended to three dimensions by collecting samples at non co-planer locations. For the UAV, a point mass model is used which is given as below:
(x_M ) ´=V_M cosa_M (1)
(y_M ) ´=V_M sina_M (2)
where, xM and yM represents the x and y coordinates of M,
VM is the velocity of the UAV 302, and aM is the heading angle of the UAV 302 with respect to the reference.
Referring back to FIG. 2, at 204, the method 200 includes collecting, during the movement of the UAV along the trajectory, a plurality of RSS samples corresponding to RF signals emitted the target. The UAV captures RSS values of the RF signal emitted by the target so as to determine the location of the target. Herein, the plurality of RSS samples are taken at equal angular separation f along the circular trajectory and corresponding to a plurality of target locations in the ROI.
At 206, the method 200 includes performing localization by using a Support Vector Regression (SVR) model, based on the plurality of RSS samples to obtain an initial estimate of target location. In an embodiment, the SVR model used for simulating the variation of RSS si (in dBm) between the UAV M and the target B is given by
s_i=s_0-10??log?_10??M-B?/d_0 +v_i,i=1,…,N (3)
where, s0 is the measured RSS at the reference distance d0,
v is the path loss factor, and vi ~ N(0; s2vi) is zero mean Gaussian distributed random variable (in dBm) with standard deviation s2vi in dBm.
The SVR model may be previously trained using a plurality of RSS training samples of the RF signals emitted by the target. In an embodiment, the SVR model is trained with a sampling pattern during an off-line stage, and learns the relationship between the RSS and the location of the UAV. The same sampling pattern that is used to train the SVR model is further utilized, during an online stage to obtain an initial estimate of location of the target.
In an embodiment, for training the SVR model, movement of the UAV is enabled along the trajectory of the sampling pattern in the ROI. Herein, the center of the sampling pattern is fixed at the center of the ROI and is taken as origin. The plurality of RSS training samples (si) corresponding to a plurality of known target locations (Bj) are collected during the movement of the UAV along the trajectory to obtain a training dataset for the SVR model. The trajectory during the training is a circular trajectory of radius R and the plurality of RSS training are samples taken at equal angular separation f and corresponding to the plurality of known target locations in the ROI. The plurality of RSS training samples (si) are provided to the SVR model for determining SVR model parameters. The SVR model determines optimal SVR model parameters (a, a*) for x co-ordinate and y co-ordinate in a dual optimization problem (as described below in equation (14)). The training of the SVR model is described below in further detail.
During the training of the SVR model, a set {sk} where k = 1…n, of RSS samples / measurements obtained by the UAV, at n sampling locations is considered. Let the aforementioned set be represented in the form of a column vector represented by s and let si represent the RSS sample vector for the ith signal source location. Let the target be located at B=(x, y). In the present embodiment, the SVR model is trained by determining a function f( f(si)) that has utmost e deviation from the actually obtained target point xi for all the training input RSS si at the same time being as flat as possible. A regression problem may be considered for approximating a linear function f that translates each input data f(si) to the output, say x-coordinate xi.
x_i=f(?("s" _i))=?("s" _i )^T "w"+b (4)

where, b and w are the parameters that needs to be learned using training data, and is an unknown constant and a unknown vector having same length as f(si), respectively.
In order to ensure that the flatness requirement is satisfied, the norm of w should be minimized, that is:
"min" 1/2||"w"||^2
subject to
x_i-?("s" _i )^T "w"-b=? (5)
x_i+?("s" _i )^T "w"+b=?
The aforementioned convex optimization problem is feasible only in cases where f exists and approximates all pairs of (si,xi) with e =0 precision, which implies convex optimization is feasible. However this may not be true always so error may have to be allowed. Introducing slack variables ?,?* to take care of the infeasible constraints of the above problem, it takes the form:
"min" 1/2?"w" ?^2+C?_(i=1)^l¦( ?_i+?_i^*) (6)

Subject to:
x_i-?("s" _i )^T "w"-b=?+?_i

x_i+?("s" _i )^T "w"+b=?+?_i^*
?_i,?_i^*=0 (7)

where, the parameter C > 0 controls the trade-off between the slack variable penalty and the flatness of f. e -intensity loss function |? |e has been described by:
|?|_?= { ¦(0,&"\ if" |?|= 3 and is stored in vector s.
The angle made by the first sampling location with the horizontal at the center of the discovery circle is represented by ?. During training, the center of the sampling pattern is fixed at the center of the ROI and is taken as the origin. Multiple RSS sample vectors si, measured at the sampling locations Mi, corresponding to different target locations Bj, j = 1, 2, 3,…, l, are collected, and fed along with the corresponding target locations Bj to the SVR model for determining the model parameters. During the training phase, the SVR model obtains the optimal parameters a and a* in Eq.(17) by maximizing Eq.(14) individually for both x and y, as given below:
¦("max" -1/2 ?_(i,j=0)¦( a_xi-a_xi^*)(a_xj-a_xj^*)k("s" _i,"s" _j)@-??_(i=1)^l¦( a_xi-a_xi^*)+?_(i=1)^l¦x_i (a_xi-a_xi^*)) (29)
¦("max" -1/2 ?_(i,j=0)¦( a_yi-a_yi^*)(a_yj-a_yj^*)k("s" _i,"s" _j)@-??_(i=1)^l¦( a_yi-a_yi^*)+?_(i=1)^l¦y_yi (a_yi-a_yi^*)) (30)
where,ax; ax* and ay; ay* are the parameters that needs to be determined for x and y coordinate to optimize Eq.(29) and (30).
FIG. 5 shows the sampling pattern with n = 5 and targets uniformly distributed over ROI with l = 5000. Since the measured RSS value is relative to the target and UAV location, this kind of setup acts as a relative frame thus giving the estimate with respect to the center of the pattern and ability to estimate the target located at any location with respect to the center of the sampling circle.
As previously discussed with reference to FIG. 2, after the training phase, during localization phase, during a movement thereof along a sampling trajectory, the UAV collects a plurality of RSS samples corresponding to RF signal emitted by the target (at 204 of method 200). Herein, the sampling trajectory followed during the localization phase is same as the one followed during the training phase in other words, the sampling trajectory includes circular trajectory of radius R, and the plurality of RSS samples are taken at equal angular separation f corresponding to the target locations spread across the ROI.
Once the UAV completes the first turn, the received data (i.e. plurality of RSS samples) are arranged with respect to the reference line, i.e., if the UAV takes a turn in counter clockwise direction, then the RSS sample that comes first after the reference line is taken as s1 (i.e. first RSS sample) and all the following samples as s2; s3,….,sn, and stored in a signal strength vector s. The angle ?’ made by a first RSS sample with the reference at the center of the circle is computed.
In an embodiment, the signal strength vector s (obtained at 204) is then provided to the trained SVR model to obtain the initial estimated location of the target as shown below in equations (31) and (32), which gives the output as the relative location of the target with respect to the center of the localization circle, but is rotated by angle - (?’- ?) as shown in FIG. 6.
x_i=?_(i=1)^l¦( a_ix-a_ix^*)k("s" _i,"s" _j)+b_x (31)
y_i=?_(i=1)^l¦( a_iy-a_iy^*)k("s" _iy,"s" _j)+b_y "\," (32)

This obtained measurement is then rotated back to the actual location using the rotation matrix as
[¦(x-() ^ y-() ^ ])=[¦(cos?(?^'-?)&-sin?(?^'-?)@sin?(?^'-?)&cos?(?^'-?))][¦(x-() ^ '@y-() ^ ')] (33)
where, (^x’, ^y’) represents the coordinates of the target obtained from SVR and (^x, ^y) shows the estimate of target location after rotation. In an embodiment, after the first turn, for each new sample taken, the latest n samples are taken and fed to the SVR model and the target location estimate is updated. An averaging filter may be used to pre-process the RSS measurements before using the RSS values for estimation. For filtering the RSS measurements, at each sampling point Mi, p number of RSS samples are collected and their mean is determined. After this only those RSS samples which satisfy equation (34) are collected and the final mean of these RSS samples is computed.
µ_s^'-t500
Value # of samples ? s µ s µ S ? s µ s
(in m) (in m) (in m) (in m) (in m) (in m) (in m) (in m) (in m) (in m)
5 14.9 13.3 27.7 20.2 51.3 37.3 84.6 59.6 123.4 90
6 14.6 12.1 29.5 21.2 49.3 35.4 79 55.8 113 82.2
7 14.2 12.5 31.8 22.4 49.3 34.4 74.4 52.6 109.2 77.9
8 16.2 19 33 23.5 49.7 34.1 72 49.4 103.4 71
9 17.2 18.1 31.1 23.7 53.8 36.2 72 49.2 106.2 72.8
10 16.2 17.2 30 22.9 55.5 36.5 72.9 49.1 107.9 72.9
11 17.3 18 28.1 21.5 55.2 36.2 74.3 48.9 107.4 70.9

From the Table II it can be seen that as the distance (r) between the localization pattern and the test point increases the mean, and the standard deviation of the error increases drastically. This is because the magnitude of the signal decreases logarithmically with distance Eq. (3), and the effect of noise at higher distances become more prominent thus increasing the estimation error. It can also be noted that the variation in the µ and t of the error for smaller r increases slightly with increase in the number of samples on the localization pattern, but at higher r values it can be seen that the µ and s decreases with increase in r. This may be because, when lesser number of sampling points are used, the distance between them is larger thus even after having lesser RSS measurements, the obtained measurements may have richer information about the sensor location as they will be less affected by the noise compared to the ones obtained when the sampling points are close to each other (though with more RSS data it will be corrupted by noise as distance between the sampling points is less and thus the measured RSS at two consecutive sampling points will be almost same). At higher radius r, the effect of noise dominates equally in both more samples and lesser samples cases, and thus the settings with larger RSS measurement samples (more data) has lower errors compared to the one with lesser numbers of samples (lesser data).
Table III: Effect on error with variation of R
r < 150 150500
R µ S ? s µ s ? s µ s
(in m) (in m) (in m) (in m) (in m) (in m) (in m) (in m) (in m) (in m)
20 19.55 15.8 40.66 29.04 75.69 52.9 123.4 82.8 184.16 134.93
40 13.85 13.27 25.07 18.54 39.32 28.15 58.56 42.59 82.7 57.61
60 14.72 15.36 19.18 19.38 28.59 20.45 39.59 28.87 59.51 37.83
80 15.77 18.01 16.63 12.91 24.22 17.61 32.33 23.33 44.31 31.59
100 13.49 14.66 14.82 12.25 20.95 15.03 26.34 18.84 37.47 26.71

Variation of radius of localization circle R: In this analysis n = 6, l = 5000 were taken. It can be seen from Table III that the general behavior of variation of error with increase in distance r from the localization pattern becomes more prominent with decrease in the radius R of the localization circle and vice versa.
Table IV: Effect on error with variation of l
r < 150 150500
l Training time(s) µ s µ s µ S µ s µ s
(in m) (in m) (in m) (in m) (in m) (in m) (in m) (in m) (in m) (in m)
3000 283.521 17.85 16.02 29.03 21.4 42.03 29.84 61.99 43.71 89 61.75
4000 523.518 15.63 14.91 25.69 18.76 39.94 28.28 60.33 42.97 86.1 60.24
5000 989.569 12.97 11.66 24.4 17.29 39.17 27.4 58.72 42.37 81.86 56.83
7000 1596.183 12.59 12.54 22.64 16.54 37.21 26.65 56.84 41.79 78.93 55.15
10000 3480.991 10.72 10.27 20.69 15.35 35.8 26.26 56.52 41.7 77.74 54.95

Variation of the size of training data l: In this analysis R = 40 and n = 6 are considered. From Table IV, it can be seen that as the number of training samples l increases, the performance of the SVR model becomes better. This is because more training data provided helps the SVR model to extract more information about the RSS and target location relationship (more support vectors), thus estimating the location with reduced error. It can also be observed that the time needed to train the SVR model increases drastically with l. Since this processing happens on more powerful workstation and is a one time job it is advantageous to use higher training samples but collecting the training samples is time consuming and laborious.
It can be seen in the above results that the mean error is very high. This is because the SVR model developed is a generalized model for the entire region of interest. Since the signal strength decreases logarithmically with increasing distance, the RSS values obtained from farther away locations are highly corrupted by the noise compared to the points closer to the sampling pattern. To take advantage of this fact, the training data was divided into two sets (R1 and R2) depending on their distance from the sampling pattern and then developing two different SVR models for individual set. Data points which are at a distance lesser than r are kept in set R1 and the once which are at distance greater then r are added to set R2 in the sampling pattern and the second one consisting of all points farther away. That is,
R1 = {Bi :|| Bi-Mc || = r} (38)
R2 = {Bi :|| Bi-Mc || = r} (39)
where, Mc represents the center of the sampling pattern. In addition to this, training data was distributed with normal distribution N(Mc,s2) rather than uniform distribution. With this approach number of training points in the R1 will be more dense, thus boosting the accuracy of the SVR model developed using R1 data set. Density of the points in R1 and R2 is controlled by varying the value of s.
Table V: Effect of s and r on accuracy
r s #R1 #R2 Error in R1 Error in R2
100 100 3935 6065 3.91 259.4
150 2007 7977 4.56 147.05
200 1110 8648 4.93 71.1
250 767 8363 6.06 64.08
300 585 7598 7.21 61.05
150 100 6803 3197 5.722 282.47
150 3862 6116 6.799 158.58
200 2448 7277 7.872 78.876
250 1618 7496 8.77 67.74
300 1206 6968 10.418 63.228
200 100 8725 1275 9.905 239.129
150 5911 4063 10.084 121.088
200 3972 5791 11.279 78.3405
250 2739 6141 12.561 66.611
300 2058 6431 13.258 63.343

Table V shows the simulation result for improved setup for three values of r and five different values of s. It can be see that with increasing value of s, mean error in R1 increases, while in R2 it decreases, this is because with increase in s, number of data points lying in R1 decreases, while in R2 it increases. Value of r < 50 was not considered as the mean error in region R2 was unacceptable similarly, for r > 200 mean error in region R1 was unacceptable.
Table V shows the simulation result for improved setup for three values of r and five different values of s. It can be see that with increasing value of s, mean error in R1 increases, while in R2 it decreases, this is because with increase in s number of data points lying in R1 decreases, while in R2 it increases. Value of r < 50 was not considered as the mean error in region R2 was unacceptable similarly, for r > 200 mean error in region R1 was unacceptable.
Herein, an implementation of the above discussed method for localizing a target using a UAV is presented. Following values have been considered: R = 30, l = 10000 and n = 6. Other parameters remains the same as in Table I. State estimation error covariance matrix is reinitialized every time when localization phase is restarted in order to improve the estimates. From table I, r = 100 and s = 200 have been chosen for the simulation. Reason behind this choice is that target lying in region R2 is always detected with mean error lesser than 100m, thus when UAV moves to this estimated approximate location and tries to localize it again, the actual location of the target is certain to lay within the region R1, which in turn will improve the accuracy further. Simulation was terminated when the variance of latest 10 target estimate becomes less than 0.01m. As already noted in the analyses on variation of R, l, and n, it can be seen in FIG. 7, that when the target is farther away from the localization pattern the target location estimation error is higher but when the UAV goes to the estimated approximate target location and continues to localize the target an estimate that has less error (or is more precise) in it is obtained.
FIG. 8 shows the variation of the error with time, the sudden increase in the error at the 400 sec represents the beginning of the 2nd localization phase where the Kalman filter is reinitialized. It can be seen that the final error in the target estimate converges very close to the actual estimate (with in 2m).
FIG. 9 is a block diagram of an exemplary computer system 901 for implementing embodiments consistent with the present disclosure. The computer system 901 may be implemented in alone or in combination of components of the system 102 (FIG. 1). Variations of computer system 901 may be used for implementing the devices included in this disclosure. Computer system 901 may comprise a central processing unit (“CPU” or “hardware processor”) 902. The hardware processor 902 may comprise at least one data processor for executing program components for executing user- or system-generated requests. The processor may include specialized processing units such as integrated system (bus) controllers, memory management control units, floating point units, graphics processing units, digital signal processing units, etc. The processor may include a microprocessor, such as AMD AthlonTM, DuronTM or OpteronTM, ARM’s application, embedded or secure processors, IBM PowerPCTM, Intel’s Core, ItaniumTM, XeonTM, CeleronTM or other line of processors, etc. The processor 902 may be implemented using mainframe, distributed processor, multi-core, parallel, grid, or other architectures. Some embodiments may utilize embedded technologies like application specific integrated circuits (ASICs), digital signal processors (DSPs), Field Programmable Gate Arrays (FPGAs), etc.
Processor 902 may be disposed in communication with one or more input/output (I/O) devices via I/O interface 903. The I/O interface 903 may employ communication protocols/methods such as, without limitation, audio, analog, digital, monoaural, RCA, stereo, IEEE-1394, serial bus, universal serial bus (USB), infrared, PS/2, BNC, coaxial, component, composite, digital visual interface (DVI), high-definition multimedia interface (HDMI), RF antennas, S-Video, VGA, IEEE 802.11 a/b/g/n/x, Bluetooth, cellular (e.g., code-division multiple access (CDMA), high-speed packet access (HSPA+), global system for mobile communications (GSM), long-term evolution (LTE), WiMax, or the like), etc.
Using the I/O interface 903, the computer system 901 may communicate with one or more I/O devices. For example, the input device 904 may be an antenna, keyboard, mouse, joystick, (infrared) remote control, camera, card reader, fax machine, dongle, biometric reader, microphone, touch screen, touchpad, trackball, sensor (e.g., accelerometer, light sensor, GPS, gyroscope, proximity sensor, or the like), stylus, scanner, storage device, transceiver, video device/source, visors, etc.
Output device 905 may be a printer, fax machine, video display (e.g., cathode ray tube (CRT), liquid crystal display (LCD), light-emitting diode (LED), plasma, or the like), audio speaker, etc. In some embodiments, a transceiver 906 may be disposed in connection with the processor 902. The transceiver may facilitate various types of wireless transmission or reception. For example, the transceiver may include an antenna operatively connected to a transceiver chip (e.g., Texas Instruments WiLink WL1283, Broadcom BCM4750IUB8, Infineon Technologies X-Gold 618-PMB9800, or the like), providing IEEE 802.11a/b/g/n, Bluetooth, FM, global positioning system (GPS), 2G/3G HSDPA/HSUPA communications, etc.
In some embodiments, the processor 902 may be disposed in communication with a communication network 908 via a network interface 907. The network interface 907 may communicate with the communication network 908. The network interface may employ connection protocols including, without limitation, direct connect, Ethernet (e.g., twisted pair 10/100/1000 Base T), transmission control protocol/internet protocol (TCP/IP), token ring, IEEE 802.11a/b/g/n/x, etc. The communication network 908 may include, without limitation, a direct interconnection, local area network (LAN), wide area network (WAN), wireless network (e.g., using Wireless Application Protocol), the Internet, etc. Using the network interface 907 and the communication network 908, the computer system 901 may communicate with devices 909 and 910. These devices may include, without limitation, personal computer(s), server(s), fax machines, printers, scanners, various mobile devices such as cellular telephones, smartphones (e.g., Apple iPhone, Blackberry, Android-based phones, etc.), tablet computers, eBook readers (Amazon Kindle, Nook, etc.), laptop computers, notebooks, gaming consoles (Microsoft Xbox, Nintendo DS, Sony PlayStation, etc.), or the like. In some embodiments, the computer system 701 may itself embody one or more of these devices.
In some embodiments, the processor 902 may be disposed in communication with one or more memory devices (e.g., RAM 913, ROM 914, etc.) via a storage interface 912. The storage interface may connect to memory devices including, without limitation, memory drives, removable disc drives, etc., employing connection protocols such as serial advanced technology attachment (SATA), integrated drive electronics (IDE), IEEE-1394, universal serial bus (USB), fiber channel, small computer systems interface (SCSI), etc. The memory drives may further include a drum, magnetic disc drive, magneto-optical drive, optical drive, redundant array of independent discs (RAID), solid-state memory devices, solid-state drives, etc. Variations of memory devices may be used for implementing, for example, any databases utilized in this disclosure.
The memory devices may store a collection of program or database components, including, without limitation, an operating system 916, user interface application 917, user/application data 918 (e.g., any data variables or data records discussed in this disclosure), etc. The operating system 916 may facilitate resource management and operation of the computer system 901. Examples of operating systems include, without limitation, Apple Macintosh OS X, Unix, Unix-like system distributions (e.g., Berkeley Software Distribution (BSD), FreeBSD, NetBSD, OpenBSD, etc.), Linux distributions (e.g., Red Hat, Ubuntu, Kubuntu, etc.), IBM OS/2, Microsoft Windows (XP, Vista/7/8, etc.), Apple iOS, Google Android, Blackberry OS, or the like. User interface 917 may facilitate display, execution, interaction, manipulation, or operation of program components through textual or graphical facilities. For example, user interfaces may provide computer interaction interface elements on a display system operatively connected to the computer system 901, such as cursors, icons, check boxes, menus, scrollers, windows, widgets, etc. Graphical user interfaces (GUIs) may be employed, including, without limitation, Apple Macintosh operating systems’ Aqua, IBM OS/2, Microsoft Windows (e.g., Aero, Metro, etc.), Unix X-Windows, web interface libraries (e.g., ActiveX, Java, Javascript, AJAX, HTML, Adobe Flash, etc.), or the like.
In some embodiments, computer system 901 may store user/application data 918, such as the data, variables, records, etc. as described in this disclosure. Such databases may be implemented as fault-tolerant, relational, scalable, secure databases such as Oracle or Sybase. Alternatively, such databases may be implemented using standardized data structures, such as an array, hash, linked list, structured text file (e.g., XML), table, or as object-oriented databases (e.g., using ObjectStore, Poet, Zope, etc.). Such databases may be consolidated or distributed, sometimes among the various computer systems discussed above in this disclosure. It is to be understood that the structure and operation of any computer or database component may be combined, consolidated, or distributed in any working combination.
Additionally, in some embodiments, the server, messaging and instructions transmitted or received may emanate from hardware, including operating system, and program code (i.e., application code) residing in a cloud implementation. Further, it should be noted that one or more of the systems and methods provided herein may be suitable for cloud-based implementation. For example, in some embodiments, some or all of the data used in the disclosed methods may be sourced from or stored on any cloud computing platform.
The written description describes the subject matter herein to enable any person skilled in the art to make and use the embodiments. The scope of the subject matter embodiments is defined by the claims and may include other modifications that occur to those skilled in the art. Such other modifications are intended to be within the scope of the claims if they have similar elements that do not differ from the literal language of the claims or if they include equivalent elements with insubstantial differences from the literal language of the claims.
Various embodiments disclosed herein provide method and system for target localization. In an embodiment, the system includes a SVR model to localize a target by using the RSS measurements of the RF signal emitted by the SVR model. In an embodiment, the sampling pattern used by the UAV includes a circular localization pattern to collect the RSS samples and same pattern is used to train the SVR model to estimate the target’s location directly. The disclosed method improves the estimation error further in the target localization. The approach used is realistic in nature and can be used in applications demanding low power consumption and cost.
It is to be understood that the scope of the protection is extended to such a program and in addition to a computer-readable means having a message therein; such computer-readable storage means contain program-code means for implementation of one or more steps of the method, when the program runs on a server or mobile device or any suitable programmable device. The hardware device can be any kind of device which can be programmed including e.g. any kind of computer like a server or a personal computer, or the like, or any combination thereof. The device may also include means which could be e.g. hardware means like e.g. an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a combination of hardware and software means, e.g. an ASIC and an FPGA, or at least one microprocessor and at least one memory with software modules located therein. Thus, the means can include both hardware means and software means. The method embodiments described herein could be implemented in hardware and software. The device may also include software means. Alternatively, the embodiments may be implemented on different hardware devices, e.g. using a plurality of CPUs.
The embodiments herein can comprise hardware and software elements. The embodiments that are implemented in software include but are not limited to, firmware, resident software, microcode, etc. The functions performed by various modules described herein may be implemented in other modules or combinations of other modules. For the purposes of this description, a computer-usable or computer readable medium can be any apparatus that can comprise, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
The illustrated steps are set out to explain the exemplary embodiments shown, and it should be anticipated that ongoing technological development will change the manner in which particular functions are performed. These examples are presented herein for purposes of illustration, and not limitation. Further, the boundaries of the functional building blocks have been arbitrarily defined herein for the convenience of the description. Alternative boundaries can be defined so long as the specified functions and relationships thereof are appropriately performed. Alternatives (including equivalents, extensions, variations, deviations, etc., of those described herein) will be apparent to persons skilled in the relevant art(s) based on the teachings contained herein. Such alternatives fall within the scope and spirit of the disclosed embodiments. Also, the words “comprising,” “having,” “containing,” and “including,” and other similar forms are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items. It must also be noted that as used herein and in the appended claims, the singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise.
Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present disclosure. A computer-readable storage medium refers to any type of physical memory on which information or data readable by a processor may be stored. Thus, a computer-readable storage medium may store instructions for execution by one or more processors, including instructions for causing the processor(s) to perform steps or stages consistent with the embodiments described herein. The term “computer-readable medium” should be understood to include tangible items and exclude carrier waves and transient signals, i.e., be non-transitory. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, nonvolatile memory, hard drives, CD ROMs, DVDs, flash drives, disks, and any other known physical storage media.
It is intended that the disclosure and examples be considered as exemplary only, with a true scope and spirit of disclosed embodiments being indicated by the following claims.