Please use this identifier to cite or link to this item:
|標題:||Multi-objective Continuous-Ant-Colony-Optimized FC forRobot Wall-Following Control||Project:||Computational Intelligence Magazine, IEEE, Volume 8, Issue3.||摘要:||
This paper proposes a multi-objective, rule-coded,advanced, continuous-ant-colony optimization (MO-RACACO)algorithm for fuzzy controller (FC) design and its application tomulti-objective, wall-following control for a mobile robot. In theMO-RACACO-based FC design approach, the number of rulesand all free parameters in each rule are optimized using the MORACACOalgorithm. This is a complex multi-objective optimizationproblem that considers both the optimization of discrete variables(number of rules) and continuous variables (rule parameters).To address this problem, the MO-RACACO uses a rule-codedindividual (solution) representation and a rule-based mutationoperation to find Pareto-optimal solutions with different numbersof rules. New solutions in the MO-RACACO are generatedusing a pheromone-level-based adaptive elite-tournament pathselection strategy followed by a Gaussian sampling operation. TheMO-RACACO-based FC design approach is applied to a multiobjective,wall-following problem for a mobile robot. Three objectivesare defined so that the robot is collision-free, maintains aconstant distance from the wall, and moves smoothly at a highspeed. This automatic design approach avoids the time-consumingmanual design of fuzzy rules and the exhaustive collection ofinput-output training pairs. The performance of the MORACACO-based control is verified through comparisons withvarious multi-objective population-based optimization algorithms(MOPOAs) in multi-objective FC optimization problems. Thisstudy also includes experiments that demonstrate robot wallfollowingcontrol using an actual mobile robot.
|Appears in Collections:||電機工程學系所|
Show full item record
TAIR Related Article
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.