Upload
vinit-payal
View
96
Download
5
Embed Size (px)
Citation preview
04/18/2023 INFORMATION FUSION 1
Submitted By :PARESH SAO(11118049)SHIVANI SINGH(11118068)TARUN BEHRA(11118079)VINIT PAYAL(11118085)
INFORMATION FUSION AND MAP BUILDING IN DISTRIBUTED SYSTEMS
VII SEMINFORMATION TECHNOLOGYNIT RAIPUR
What Is Fusion ?Fusion Is the process of the joining two or more things together to form a single entity .
Two Types of fusions in the field of computer science:-
1. Data Fusion
2. Information Fusion
Data Fusion:-
Data fusion is used for raw data (obtained directly from the sensors)
Information Fusion:-
Information fusion is employed to define already processed data.
Information fusion implies a higher semantic level than data fusion.
Data or information to be fused can be from two sources:-
1. Sensors
2. Databases The goal of using data/Information fusion in multi sensor environments is to obtain a lower detection error probability and a higher reliability by using data from multiple sources.
Information Fusion is the act or process of combining or associating data or information from one or more sources to obtain improved information for detection, identification, or characterization of that entity.
Information Fusion:-
User node
CommunicationCommunication
Sourcenode
Data from Source
Sourcenode
Fuseddata
Source node
Classification of Data Fusion Techniques
1. Classification Based on the Relations between the Data Sources:-
a) Complementary:-
when the information provided by the input sources represents different parts of the scene and could thus be used to obtain more complete global information. For example, in the case of visual sensor networks, the information on the same target provided by two cameras with different fields of view is considered complementary.
b) Redundant:-
when two or more input sources provide information about the same target and could thus be fused to increment the confidence. For example, the data coming from overlapped areas in visual sensor networks are considered redundant.
c) cooperative:- when the provided information is combined into new information that is typically more complex than the original information. For example, multi-modal (audio and video) data fusion is considered cooperative.
2. Classification Based on the Abstraction Levels:-A. low level fusionB. medium level fusion C. high level fusionD. multiple level fusion
A. low level fusion:- The raw data are directly provided as an input to the data fusion process, which provide more accurate data (a lower signal-to-noise ratio) than the individual sources;
B. Medium level:-
Characteristics or features (shape, texture, and position) are fused to obtain features that could be employed for other tasks. This level is also known as the feature or characteristic level
C. High Level:-
This level, which is also known as decision fusion, takes symbolic representations as sources and combines them to obtain a more accurate decision. Bayesian’s methods are typically employed at this level
D. Multiple level fusion:
This level addresses data provided from different levels of abstraction (i.e., when a measurement is combined with a feature to obtain a decision) :-
3. Dasarathy’s Classification:-
A. data in-data out (DAI-DAO)
B. data in-feature out (DAI-FEO)
C. feature in-feature out (FEI-FEO)
D. feature in-decision out (FEI-DEO)
E. Decision In-Decision Out (DEI-DEO)
A. Data in-data out (DAI-DAO):-
This type of data fusion process inputs and outputs raw data; the results are typically more reliable or accurate. Data fusion at this level is conducted immediately after the data are gathered from the sensors. The algorithms employed at this level are based on signal and image processing algorithms .
B. data in-feature out (DAI-FEO):-
At this level, the data fusion process employs raw data from the sources to extract features or characteristics that describe an entity in the environment .
C. feature in-feature out (FEI-FEO):-
At this level, both the input and output of the data fusion process are features. Thus, the data fusion process addresses a set of features with to improve, refine or obtain new features. This process is also known as feature fusion, symbolic fusion, information fusion or intermediate-level fusion
D. feature in-decision out (FEI-DEO):-
This level obtains a set of features as input and provides a set of decisions as output. Most of the classification systems that perform a decision based on a sensor’s inputs fall into this category of classification.
E. Decision In-Decision Out (DEI-DEO):-
This type of classification is also known as decision fusion. It fuses input decisions to obtain better or new decisions.
04/18/2023 INFORMATION FUSION 11
INFORMATION FUSION ARCHITECTURE
There are three types of Information fusion Architecture
•Centralized•Decentralize•Hierarchal
04/18/2023 INFORMATION FUSION 12
CENTRALIZED
• It is the Simplest of all the architecture.• In this a central processor fuses the reports collected by all other sensing
nodes i.e. the data is collected from different nodes and a centralized node produces the final output.
Advantage: Erroneous report(s) can be easily detected.
Disadvantage: Inflexible to sensor changes and the workload is concentrated at a single point.
04/18/2023 INFORMATION FUSION 13
CENTRALIZED FUSION
Arch. Type : Centralized Processing : Moderate Bandwidth : Moderate Timeliness : Poor Central
node
Sensor measurement
data
04/18/2023 INFORMATION FUSION 14
DECENTRALIZED
• Data fusion occurs locally at each node on the basis of local observations and the information obtained from neighboring nodes. • No central processor node.
Advantages:Scalable and tolerant to the addition or loss of sensing nodes or dynamic changes in the network.
04/18/2023 INFORMATION FUSION 15
HIERARCHICAL
• Nodes are partitioned into hierarchical levels.• The sensing nodes are at level 0 and the BS at the highest level.• Reports move from the lower levels to higher ones.
Advantage:Workload is balanced among nodes
04/18/2023 INFORMATION FUSION 16
HIERARCHICAL INFORMATION FUSION ARCHITECTURE
Arch. Type : Hierarchical Processing : High at mid levels Bandwidth : Moderate Timeliness : Good
Fusion Node
Sensor data 1
Sensor data 2
Sensor data 3
Sensor data 1+2
Sensor data 1+3
04/18/2023 INFORMATION FUSION 17
BENEFITS FROM INFORMATION FUSION SYSTEM
•Fusion process is necessary most of all to reduce (to filter) input information through its integration (merging) and generalization.
•Fusion process is necessary to improve accuracy.
Fusion process is necessary to reduce uncertainty.
04/18/2023 INFORMATION FUSION 18
INFORMATION FUSION APPLICATION
MILITARY APPLICATION :•Location and characterization of enemy units and weapons•High level inferences about enemy situation•Air to air or surface to air defense•Ocean monitoring•Battlefield intelligence•Strategic warning
NON MILITARY APPLICATIONS:•Condition based maintenance•Detection of system faults•Robotics•Medical•Environmental monitoring•Location and identification of natural phenomena
04/18/2023 INFORMATION FUSION 19
Coding Part
Node 1Turtle_bot
Node 2Navigation Ser.
Use data for
robot navigation
Robot
navigation
Send /scan data
Send /cmd_vel_nev Map
Generation
Processed /cmd_vel_nevWithSLAM
04/18/2023 INFORMATION FUSION 20
REFERNCES Andersson, L. A. (2008). Multi-robot Information Fusion. Sweden: UniTryck, Link¨oping.
Arwin Datumaya, W. S. (2008). Design and Implementation of Multi Agent-based information Fusion System for Decision Making Support.
Kiril Alexiev, I. N. (2006). Methods for Data and Information Fusion .
Xiang Li, M. S. (2009). Autonomous Information Fusion for Robust Obstacle Localization on a Humanoid Robot. The Latin American Robotics Symposium .