Combining robot control strategies using genetic algorithms with memory

Author(s):  
Sushil J. Louis ◽  
Gan Li
2008 ◽  
Vol 16 (3) ◽  
pp. 385-416 ◽  
Author(s):  
Shengxiang Yang

In recent years the genetic algorithm community has shown a growing interest in studying dynamic optimization problems. Several approaches have been devised. The random immigrants and memory schemes are two major ones. The random immigrants scheme addresses dynamic environments by maintaining the population diversity while the memory scheme aims to adapt genetic algorithms quickly to new environments by reusing historical information. This paper investigates a hybrid memory and random immigrants scheme, called memory-based immigrants, and a hybrid elitism and random immigrants scheme, called elitism-based immigrants, for genetic algorithms in dynamic environments. In these schemes, the best individual from memory or the elite from the previous generation is retrieved as the base to create immigrants into the population by mutation. This way, not only can diversity be maintained but it is done more efficiently to adapt genetic algorithms to the current environment. Based on a series of systematically constructed dynamic problems, experiments are carried out to compare genetic algorithms with the memory-based and elitism-based immigrants schemes against genetic algorithms with traditional memory and random immigrants schemes and a hybrid memory and multi-population scheme. The sensitivity analysis regarding some key parameters is also carried out. Experimental results show that the memory-based and elitism-based immigrants schemes efficiently improve the performance of genetic algorithms in dynamic environments.


2015 ◽  
pp. 787-817
Author(s):  
Saeid Pourzeynali ◽  
Shide Salimi

The main objective of this chapter is to find the optimal values of the parameters of the base isolation systems and that of the semi-active viscous dampers using genetic algorithms (GAs) and fuzzy logic in order to simultaneously minimize the buildings' selected responses such as displacement of the top story, base shear, and so on. In this study, performance of base isolation systems, and semi-active viscous dampers are studied separately as different vibration control strategies. In order to simultaneously minimize the objective functions, a fast and elitist non-dominated sorting genetic algorithm (NSGA-II) approach is used to find a set of Pareto-optimal solution. To study the performance of semi-active viscous dampers, the torsional effects exist in the building due to irregularities, and unsymmetrical placement of the dampers is taken into account through 3D modeling of the building.


1990 ◽  
Vol 10 (2) ◽  
pp. 16-22 ◽  
Author(s):  
M. Buhler ◽  
D.E. Koditschek ◽  
P.J. Kindlmann

Author(s):  
D J Brookfield

One of the main difficulties in introducing improved robot control strategies is a lack of knowledge of the frictional behaviour of robot drive systems. The aim of the present paper is to describe a technique for the identification of Coulomb friction based on the response of the robot drive to a sinusoidal driving torque The presence of a third harmonic component in the resulting velocity is a consequence of the Coulomb non-linearity and it is shown theoretically, through computer simulation and in experimental tests, that the coefficient of Coulomb friction can be estimated from the amplitude of the third harmonic component. The identification method is shown to be applicable to any mechanical system that can be subjected to a sinusoidal forcing torque or force.


1995 ◽  
Vol 52 (9) ◽  
pp. 1893-1908 ◽  
Author(s):  
Donna J. D'Angelo ◽  
Judy L. Meyer ◽  
Leslie M. Howard ◽  
Stanley V. Gregory ◽  
Linda R. Ashkenas

Genetic algorithms (GA) are artificial intelligence techniques based on the theory of evolution that through the process of natural selection evolve formulae to solve problems or develop control strategies. We designed a GA to examine relationships between stream physical characteristics and trout distribution data for 3rd-, 5th-, and 7th-order stream sites in the Cascade Mountains, Oregon. Although traditional multivariate statistical techniques can perform this particular task, GAs are not constrained by assumptions of independence and linearity and therefore provide a useful alternative. To help gauge the effectiveness of the GA, we compared GA results with results from proportional trout distributions and multiple linear regression equations. The GA was a more effective predictor of trout distributions (paired t test, P < 0.05) than other methods and also provided new insights into relationships between stream geomorphology and trout distributions. Most importantly, GA equations emphasized the nonindependence of stream channel units by revealing that (i) the factors that influence trout distributions change along a downstream continuum, and (ii) channel unit sequence can be critical. Superior performance of the GA, along with the new information it provided, indicates that genetic algorithms may provide a useful alternative or supportive method to statistical techniques.


Author(s):  
Luiz Carlos Figueiredo ◽  
Gilcésar Ávila ◽  
Francisco Reinaldo ◽  
Rui Camacho ◽  
Demétrio R. Magalhães ◽  
...  

In this paper we report as the development of a tool in to develop and set control strategies as a fast and easy way. Additionally, a tricycle robot with two traction motors was built to test the strategies produced with the tool. Experimental tests have shown an advantage in the use of such tool.


Sign in / Sign up

Export Citation Format

Share Document