Behavior coordination of autonomous mobile robot navigation by neuro-fuzzy system

Khatoon, Shahida

doi:10.1109/nebc.2005.1431924

Cited by 6 publications

(12 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, a common problem arises in command fusion techniques, when competing behaviors issue conflicting control commands resulting in local minima or stagnant situations of the robot [9,46]. This problem is addressed by weighted decision-making techniques [47,48,49,50,51,52,53,54,55], where different sets of rules are used to weight the conflicting commands according to environmental contexts, thus, reducing the possibility of producing local minima.…”

Section: Problem I: Behavior Selectionmentioning

confidence: 99%

“…A behavior coordinator selects appropriate behaviors for a given world state. The available knowledge representation techniques include Finite State Automata (FSA) [35,36,39], Petri Net [16], decision-tree [13,24], and fuzzy-rule [48,49,50,51,52,53,54,55,56] based approaches. These approaches mainly differ from each other depending on the use of binary or multi-valued logic to construct the context rules.…”

Section: Problem Ii: Knowledge Representationmentioning

confidence: 99%

“…If the system does not preserve previous activation states [11,14,18,23,33,37,38,40,52,55,66,68,77], it is characterized as without memory.…”

Section: The Proposed Classificationmentioning

confidence: 99%

“…Finally, the top-most behavior generates the motion commands. Single level approaches [11,13,14,34,37,38,40,42,55,61,62,63,66,77] do not utilize a layered behavioral decomposition; instead prior knowledge and current sensory feedback are directly used to select a subset of primitive behaviors.…”

Section: The Proposed Classificationmentioning

confidence: 99%

“…• Off-line learning: These techniques [55,63] employ off-line learning approaches to train a Neural Network (NN) that determines the current world state using the sensory data and then selects an appropriate set of behaviors to control the robot.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Behavior-based Robot Control Using Fuzzy Discrete Event System

Huq

Mann

Gosine

2006

2006 IEEE International Conference on Fuzzy Systems

View full text Add to dashboard Cite

The key issue associated with behavior-based architecture is the development of a behavior coordinator. A behavior coordinator should possess a number of important properties. The literature suggests that (a) both behavior arbitration and command fusion techniques should be combined in order to coordinate competitive and cooperative behaviors, (b) the coordinator should possess adequate means of modeling the current state of the world, (c) sensory uncertainties should be modeled using multivalued logic, (d) the coordinator should be capable of making persistent behavior selection, (e) it should provide the opportunity to accommodate hierarchical decisionmaking for reactive action generation, (f) it should possess predictive decision-making capability for handling future environmental uncertainties, (g) it should provide satisfactory means of decision analysis, and finally (h) it should be modular to achieve robustness with a larger number of behaviors. This thesis attempts to address some of the issues mentioned above while exploiting fuzzy logic (FL) based methodologies for behavior coordination.The initial investigation includes experimentation of different methodologies in the literature. A FL-based controller is designed for motor schema based mobile robot navigation. Fuzzy meta rules have been used to adaptively generate weights for each motor schema. It uses human reasoning to model the deterministic uncertainty of sensory data, which in turn helps to reduce the possibility of generating incorrect weights for motor schemas. Experimental results demonstrate that the fuzzy logic based approach overcomes some of the common problems of schema-based navigation.It has been observed that a pure FL-based method has several disadvantages, such as it imposes a scalability problem when the system consists of both competitive and cooperative behaviors and it does not provide any feedback measures for decision analysis. Moreover, the rule-based knowledge representation becomes complex in cases where previous state information is incorporated for persistent decision-making.While understanding the shortcomings of each technique, this thesis present a novel behavior coordination architecture using Fuzzy Discrete Event System (FDES).This architecture addresses the shortcomings of the FL-based technique using complimentary properties of Discrete Event System (DES). The DES-based method provides supervisory control techniques for behavior arbitration and has a suitable framework for decision analysis in terms of observability and controllability. Furthermore, it supports multi-level behavioral decomposition and uses previous state information of the system for persistent decision-making. As a result, the combination of FL and DES provides the opportunity to integrate several key features to achieve a high performance behavior coordinator.Finally the proposed architecture is experimentally tested using three different robotic applications, namely robotic navigation, robotic box pulling and robotic visual attention. Th...

show abstract

Section: Problem I: Behavior Selectionmentioning

confidence: 99%

Section: Problem Ii: Knowledge Representationmentioning

confidence: 99%

“…If the system does not preserve previous activation states [11,14,18,23,33,37,38,40,52,55,66,68,77], it is characterized as without memory.…”

Section: The Proposed Classificationmentioning

confidence: 99%

Section: The Proposed Classificationmentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

Behavior-based Robot Control Using Fuzzy Discrete Event System

Huq

Mann

Gosine

2006

2006 IEEE International Conference on Fuzzy Systems

View full text Add to dashboard Cite

show abstract

PCA-based method to fuse behaviors from place characterization for robot navigation

Poncela

2015

Robotica

View full text Add to dashboard Cite

SUMMARYThis paper presents a method to calculate the fusing rule among three reactive behaviors, Wall Following, Corridor Following and Door Crossing, from place characterization for robot navigation. The technique is supported by a local grid of the closest area to the robot, which is built from sonar readings. The contour of this grid is extracted, represented by itsFFTand, finally, it is reduced to a short feature vector with a principal component analysis (PCA). This feature vector is used to decide the fusing rule among the three behaviors. The algorithm is very fast in terms of its time performance, being then valid to be used in robot navigation, since the robot would rapidly react to new situations. It has also been successfully tested in simulated and real environments, with a Pioneer robot equipped with eight frontal sonar sensors, both in manually driven tasks and autonomous navigation tasks, proving its feasibility and effectiveness.

show abstract

Behavior Fusion of Robot Navigation Using a Fuzzy Neural Network

Song

Lin

2006

2006 IEEE International Conference on Systems, Man and Cybernetics

View full text Add to dashboard Cite

This paper presents a design of behaviorfusion architecture for mobile robot navigation. We first design three behaviors for robot navigation, including obstacle avoidance, wall following, and goal seeking. We implement these primitive behaviors by using fuzzy-logic control approaches. Then, the fusion weight of each behavior is determined by using the proposed behaviorfusion neural network. The neural network maps the current environment sensor data to suitable fusion weights. Both computer simulation and practical experiments verify the effectiveness of the method.

show abstract

Behavior coordination of autonomous mobile robot navigation by neuro-fuzzy system

Cited by 6 publications

References 28 publications

Behavior-based Robot Control Using Fuzzy Discrete Event System

Behavior-based Robot Control Using Fuzzy Discrete Event System

PCA-based method to fuse behaviors from place characterization for robot navigation

Behavior Fusion of Robot Navigation Using a Fuzzy Neural Network

Contact Info

Product

Resources

About