Behavior-based learning fuzzy rules for mobile robots

Thongchai, Siripun

doi:10.1109/acc.2002.1023148

Cited by 16 publications

(10 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The algorithm has been demonstrated to build a robot controller capable of performing obstacle avoidance in a realworld environment (Thongchai 2002) and the algorithm has been shown to cope effectively with limited noise (Carmona et al 2004). Fully understanding the implications of noise is important for a control problem with a continuous state space.…”

Section: Wang and Mendelmentioning

confidence: 97%

Learning a robot controller using an adaptive hierarchical fuzzy rule-based system

Waldock

Carse

2015

Soft Comput

View full text Add to dashboard Cite

The majority of machine learning techniques applied to learning a robot controller generalise over either a uniform or pre-defined representation that is selected by a human designer. The approach taken in this paper is to reduce the reliance on the human designer by adapting the representation to improve the generalisation during the learning process. An extension of a Hierarchical Fuzzy Rule-Based System (HFRBS) is proposed that identifies and refines inaccurate regions of a fuzzy controller, while interacting with the environment, for both supervised and reinforcement learning problems. The paper shows that a controller using an adaptive HFRBS can learn a suitable control policy using a fewer number of fuzzy rules for both a supervised and reinforcement learning problem and is not sensitive to the layout as with a uniform representation. In supervised learning problems, a small number of extra trials are required to find an effective representation but for reinforcement learning problems, the process of adapting the representation is shown to significantly reduce the time taken to learn a suitable control policy and hence open the door to high-dimensional problems.

show abstract

Section: Wang and Mendelmentioning

confidence: 97%

Learning a robot controller using an adaptive hierarchical fuzzy rule-based system

Waldock

Carse

2015

Soft Comput

View full text Add to dashboard Cite

show abstract

“…Remember that every real value is transformed For the inference process (in the defuzzification) we used an in two fuzzy-sets by the fuzzification. In case of the rule output adaptation of the minimum and product classic method [2], [5], is like one of the two sets of the user decision the system will [10], [1], [8], [7], with the consideration of the weights assigned increase the weight by the reason of the fourth part of the set at each fuzzy-rules. This idea will be explained in the next membership value.…”

Section: User Adaptive Learning Algorithmmentioning

confidence: 99%

Fuzzy Logic User Adaptive Navigation Control System For Mobile Robots In Unknown Environments

Olivares-Méndez

2007

2007 IEEE International Symposium on Intelligent Signal Processing

View full text Add to dashboard Cite

-This paper presents a software implementation of a "fuzzy sets" was 1965 by Lofti A. Zadeh. Since then, the use of user adaptive fuzzy control system for autonomous navigation in fuzzy logic techniques in the robotic field has become a common mobile robots for unknown environments. This system has been approach. tested in a pioneer mobile robot and on a robotic wheelchair, fitted Ir with PLS laser sensor to detect the obstacles and odometry sensorsIn the following paper we propose a set of improvements for localization of robots and the goal positions. The system is like, a more comprehensive sector for the definition of each able to drive the robots to their goal position avoiding static and fuzzy-variable and a user adaptive learning algorithm using the dynamic obstacles, without using any pre-built map. Our approach synapses-weight idea of the brain operation. For this purpose learn from user behaviors in the way it can resolve different situations against obstacles or walls.We propose and implement we createavrti ftware caledMOs (iu liares two updates for the fuzzy system. For the implementation of the Fuzzy Software) with fuzzy logic techniques that can learn from learning algorithm we use a weighting scheme giving a value for experience regardless the intelligent system architecture.each fuzzy-rule, this value is based on the synapse-weight idea and In the next section of this paper the autonomous fuzzy represent the contribution of each rule in the system output. We navigation control system is detailed. Section III define the also create of a more important sector in the definition of the fuzzyav ation cont sysem is detaled.ISectionId variables, based on a statistics system that measure the uses of all u the sets of the variables in order to contract the size of the rule-base. describes the fuzzy-software implementation. Section V show the experimental results. Section VI conclusion and future works. KeVwords -Fuzzy logic, autonomous navigation, adaptive control, fuzzy control, mobile robots, collision avoidance, reactive naviga-1 AUTONOMOUs Fuzzy NAvIGATION CONTROL SYSTEM tion In order to express in a more flexible way our approach we I. INTRODUCTION have developed an autonomous fuzzy navigation system. The structure of the fuzzy control is divided in three inputs variables The autonomous navigation of mobile robots is one of the for interaction with the environment and one output. The data we most investigated problems in the robotic community. The use to check each situation is defined by the distance (in meters) usual way to face this problem is with reactive, delivered between the robot and the near obstacle, the angle direction or hybrid methods. One of the biggest problem on those between obstacles and the angle to the goal from the robot systems are the computational and memory cost. We approach position (in degrees). The angles and distance are recovered the computational cost avoiding and reducing the use of using a 180 range laser sensor and an odometry system. The environment maps...

show abstract

“…And, what kind of fuzzy controller is the most adequate? The most popular groups of learning methodologies for fuzzy controllers in robotics are evolutionary algorithms (Dahl & Giraud-Carrier, 2004;Gu, Hu, Reynolds, & Tsang, 2003;Hagras, Callaghan, & Collin, 2004;Izumi, Watanabe, & Jin, 1999;Katagami & Yamada, 2000;Mucientes, Moreno, Bugarín, & Barro, 2006;Mucientes, Moreno, Bugarín, & Barro, 2007;Oh & Barlow, 2004;Yamada, 2005), neural networks (Hui, Mahendar, & Pratihar, 2006;Shiah & Young, 2004) and reinforcement learning (Beom & Cho, 1995;Bonarini, 1997;Gu, Hu, & Spacek, 2003;Kalmár, Szepesvári, & Lörincz, 1998;Lin, 2003;Takahashi & Asada, 2003;Thongchai, 2002;Wang, Huber, Papudesi, & Cook, 2003;Zhou, 2002). Also, combinations of them, like neural networks and evolutionary algorithms (Berlanga, Sanchis, Isasi, & Molina, 2000;Chen & Chiang, 2004;Floreano & Mondada, 1998;Lee & Zhang, 2000;Miglino, Lund, & Nolfi, 1995;Nelson, Grant, Barlow, & White, 2003;Tuci, Quinn, & Harvey, 2003), have been successfully applied.…”

Section: Introductionmentioning

confidence: 98%

A case study for learning behaviors in mobile robotics by evolutionary fuzzy systems

Mucientes

Alcalá‐Fdez

Alcalá

et al. 2010

Expert Systems with Applications

View full text Add to dashboard Cite

Behavior-based learning fuzzy rules for mobile robots

Cited by 16 publications

References 9 publications

Learning a robot controller using an adaptive hierarchical fuzzy rule-based system

Learning a robot controller using an adaptive hierarchical fuzzy rule-based system

Fuzzy Logic User Adaptive Navigation Control System For Mobile Robots In Unknown Environments

A case study for learning behaviors in mobile robotics by evolutionary fuzzy systems

Contact Info

Product

Resources

About