An intelligent control system with a multi-objective self-exploration process

doi:10.1016/S0165-0114(03)00183-0

Fuzzy Sets and Systems

Volume 143, Issue 2, 16 April 2004, Pages 275-294

https://doi.org/10.1016/S0165-0114(03)00183-0 Get rights and content

Abstract

This paper proposes a novel approach based on artificial intelligence technologies (multi-objective Self-Exploration process based Intelligent Control System—mSEICS) for intelligent control systems. Not only can this system adapt to various environments, but it can also continually improve its performance. The mSEICS consists of four basic functions, controller, receptor, m-adaptor and advancer. A five-layer fuzzy neural network is applied to implement the controller. The receptor is used to evaluate the performance of system. The m-adaptor (multi-objective based adaptor) that comprises two elements, action explorer and rule generator, can generate a variety of new action sets in order to adapt to various environments. The Pareto optimality based multi-objective genetic algorithm is proposed to implement the action explorer to discover multiple action sets, and the rule generator is employed to transform the action set to fuzzy rules. In addition, the advancer consisting of action discoverer and rule generator is constructed to produce the novel action set to enhance the system efficiency. The parallel-simulated annealing approach is presented to realize the action discoverer. An application of the robotic path planning is applied to demonstrate the proposed model. The simulation results show that the mobile robot can reach the target successfully in various environments, and the proposed model is more efficient than the similar model.

References (33)

M.-R. Akbarzadeh-T et al.
Soft computing for autonomous robotic systems
Comput. Electric. Eng.
(2000)
J.S. Albus
The engineering of mind
Inform. Sci.
(1999)
J. Dean
Animats and what they can tell us
Trends Cognitive Sci.
(1998)
T. Fukuda et al.
Intelligent systemsrobotics versus mechatronics
Annual Rev. Control
(1998)
B. Hayes-Roth
Integrating real-time AI techniques in adaptive intelligent agents
Annual Rev. Automat. Programming
(1994)
B. Hayes-Roth
An architecture for adaptive intelligent systems
Artificial Intelligence
(1995)
J. Kim et al.
HyFISadaptive neuro-fuzzy inference systems and their application to nonlinear dynamical systems
Neural Networks
(1999)
D.A. Linkens et al.
Intelligent control of a cryogenic cooling plant based on blackboard system architecture
ISA Trans.
(2000)
T. Morimoto et al.
An intelligent control for greenhouse automation, oriented by the concepts of SPA and SFA—An application to a post-harvest process
Comput. Electron. Agric.
(2000)
T. Morimoto et al.
AI approaches to identification and control of total plant production systems
Control Eng. Practice
(2000)

T. Morimoto et al.

Optimization of heat treatment for fruit during storage using neural networks and genetic algorithms

Comput. Electron. Agric.

(1997)

J.J. Shann et al.

A fuzzy neural network for rule acquiring on fuzzy control system

Fuzzy Sets and Systems

(1995)

T. Shibata et al.

Skill based motion planning in hierarchical intelligent control of a redundant manipulator

Robotics Autonomous Systems

(1996)

C.D. Stylios et al.

Fuzzy cognitive mapsa model for intelligent supervisory control systems

Comput. Indust.

(1999)

J.S. Albus

Outline for a theory of intelligence

IEEE Trans. Systems Man Cybernet.

(1991)

L.-H. Chen et al.

New approach to intelligent control systems with self-exploring process

IEEE Trans. Systems Man Cybernet. Part B: Cybernet.

(2003)

Cited by (13)

A case study for learning behaviors in mobile robotics by evolutionary fuzzy systems
2010, Expert Systems with Applications
Service robots will play an increasing and more important role in the society in the next years. One of the main challenges is to endow robots with enough autonomy to operate on real environments. To reach that goal, the design of controllers to solve simple tasks must be automatized. Engineers look for learning algorithms that are general, robust, require low expertise knowledge, and generate controllers that can run on the real robot without any tuning stage. In this paper, a framework to learn behaviors (controllers) in mobile robotics, fulfilling the previous requirements, has been used. The framework is based on two modules: dataset generation and a data-driven evolutionary-based learning algorithm to obtain fuzzy controllers. Nevertheless, the design of a fuzzy controller still requires the selection of the type of learning algorithm, and also to choose the value of some design parameters. In this paper we present an exhaustive study on a set of evolutionary-based data-driven learning algorithms, for learning fuzzy controllers in mobile robotics, that cover a wide range of the accuracy/interpretability trade-off. The study has also evaluated the influence of the values of all the design parameters over accuracy and interpretability. The objective is to analyze the performance of the different algorithms for the design of behaviors in mobile robotics, and to extract some general rules that can help in the process to design new behaviors. The analysis comprises two different behaviors (wall-following and moving object following) and more than 450 tests, both in simulation and on a Pioneer II AT robot. Results have shown very good performances in complex and realistic conditions for the different combinations of algorithms and parameters.
A symbol-based intelligent control system with self-exploration process
2008, Engineering Applications of Artificial Intelligence
Citation Excerpt :
We compare the performance of the SyICS approach with that of the SEICS (Chen and Chiang, 2003) and mSEICS (Chen and Chiang, 2004) models. The purpose of robotic path planning is to obtain the shortest path for the robot from a start point to a target point without collisions (Chen and Chiang, 2003, 2004; Cosio and Castaneda, 2004; Nearchou, 1999). In this paper, we attempt to generate the approximate shortest path without collisions to demonstrate the adaptability of our model.
This paper presents a symbol-based intelligent control system (SyICS) with a self-exploration process. The SyICS is comprised of a symbolic controller, a percepter, and a self-adaptor, and is a rule-based control system with on-line parametric adaptation. The symbolic controller consists of a number of symbolic rules, such as IF–THEN rules, for controlling the plant. The percepter is a sensory mechanism to perceive the control efficiency. Once the sensory information is found to be improper, i.e., there is inefficient control, the self-adaptor will be activated; otherwise, the symbolic controller will keep on the controlling assignment. The self-adaptor is an adaptive mechanism to explore the new symbolic rules and update the knowledge base for on-line and real-time adaptation. The self-exploration process is applied for the self-adaptor, and the hybrid genetic algorithm with variable-length chromosome is presented to fulfill the self-exploration process. The advantages of the SyICS are: (1) the symbolic controller is intuitive and easy to implement, and (2) The mechanism of the on-line adaptation is adopted and performed by the efficient hybrid genetic algorithm. A robotic path planning application is used to demonstrate the SyICS approach by comparing it with other intelligent control methods. The simulation results show that the robotic paths of SyICS model are the most efficient for all cases based on the path's efficiency measure.
A self-adaptive intelligent control system with hierarchical architecture
2015, Journal of Information Science and Engineering
Multiobjective genetic fuzzy systems
2015, Springer Handbook of Computational Intelligence
Research and design of embedded intelligent monitoring system for plant factory
2014, Applied Mechanics and Materials
Research on the speed controlling of continuously spinning north seeker based on fuzzy self-adaptor PID
2012, Journal of Convergence Information Technology

View all citing articles on Scopus

View full text

An intelligent control system with a multi-objective self-exploration process

Abstract

Comput. Electric. Eng.

Inform. Sci.

Trends Cognitive Sci.

Annual Rev. Control

Annual Rev. Automat. Programming

Artificial Intelligence

Neural Networks

ISA Trans.

Comput. Electron. Agric.

Control Eng. Practice

Comput. Electron. Agric.

Fuzzy Sets and Systems

Robotics Autonomous Systems

Comput. Indust.

Outline for a theory of intelligence

IEEE Trans. Systems Man Cybernet.

New approach to intelligent control systems with self-exploring process

IEEE Trans. Systems Man Cybernet. Part B: Cybernet.