Natural Motion Generation of Biped Locomotion Robot using Hierarchical Evolutionary Algorithm in the Various Environments

1997 ◽  
Vol 9 (6) ◽  
pp. 496-502
Author(s):  
Takemasa Arakawa ◽  
◽  
Toshio Fukuda
Keyword(s):  
Evolutionary Algorithm ◽  
Energy Minimization ◽  
Dynamic Effect ◽  
The Other ◽  
Human Walking ◽  
Motion Trajectory ◽  
Biped Locomotion ◽  
Motion Generation ◽  
Natural Motion ◽  
Generation Problem

The purpose of this research is to generate natural motion in a biped locomotion robot, like a human walking in various environments. In this paper, we report on biped locomotion robots. We apply a hierarchical evolutionary algorithm in order to generate the trajectory of a biped locomotion robot through energy optimization, and attempt to generate a more natural motion by considering the dynamic effect. The hierarchical evolutionary algorithm consists of two layers: one is the GA layer which minimizes the total energy of all the actuators, and the other is the EP layer which optimizes the interpolated configuration of the biped locomotion robot. Then we formulate a trajectory generation problem as an energy minimization problem and we apply the hierarchical evolutionary algorithm. Furthermore, we build a trial biped locomotion robot which has 13 joints and is made of aluminum. Finally, we confirm that the calculated natural motion trajectory can be applied successfully to practical biped locomotion.

A Novel Synthesis Method for Three-Position Motion Generation With Planar Four-Bar Mechanisms

2012 ◽  
Author(s):  
Jianyou Han ◽  
Tong Yang
Keyword(s):  
Synthesis Method ◽  
Solution Space ◽  
The Other ◽  
Motion Generation ◽  
Numerical Examples ◽  
Novel Synthesis ◽  
Generation Problem

This paper deals with the three-position motion generation problem with two specific grounded link lengths. There are two infinities of solutions for selecting the two links on the derived contours of the link lengths. These points on the contours are circle points or center points. After one half of the basic four bar had been selected on the contour, two infinities of solutions remained. These solutions can be mapped in a plane to determine where the particular types of mechanisms occur. Furthermore after one half of the basic four bar had been selected on the contour, one infinity of solutions still remained on the other contour. This indicates two infinities of solutions are still remained for the two given link lengths. These contours can be displayed in the solution space in which the motion generation is defined. With these significant useful information the better mechanism can be obtained, which satisfies more design conditions. Expressions of the contours are derived. Two numerical examples are used for illustration, but the results can be applied to any three-position motion generation problem.

Motion Trajectory Simulation and Measurement of the 5000V Electromagnetic Repulsion Mechanism (ERM)

2012 ◽  
Vol 546-547 ◽  
pp. 368-373
Author(s):  
Jin Wu ◽  
Jin Wu Zhuang ◽  
Yong Hua Zhuang ◽  
Jun Lu
Keyword(s):  
Comparative Analysis ◽  
Simulation Models ◽  
Displacement Transducer ◽  
Linear Displacement ◽  
The Other ◽  
Displacement Sensor ◽  
Motion Trajectory ◽  
Motion Trajectories ◽  
Trajectory Simulation ◽  
Piezoelectric Acceleration

Successful applications of the ERM depend on accurate simulation models. Firstly, we built up a model on the 5000 volts ERM with ansoft. Then we measured motion trajectories of the ERM prototype respectively by means of a piezoelectric acceleration transducer, a magnetic grating displacement transducer and a linear displacement sensor. Based on the comparative analysis of the results from simulation and measurement, we concluded that the simulation model was reliable on one hand, and on the other hand by means of a linear displacement sensor along with a magnetic grating counterpart, we could obtain accurate motion trajectories with a relative error less than 5%.

scholarly journals Comparison of synchronous and asynchronous parallelization of extreme surrogate-assisted multi-objective evolutionary algorithm

2020 ◽  
Author(s):  
Tomohiro Harada ◽  
Misaki Kaidan ◽  
Ruck Thawonmas
Keyword(s):  
Machine Learning ◽  
Evolutionary Algorithm ◽  
Optimization Problems ◽  
Computing Time ◽  
The Other ◽  
Multi Objective Optimization ◽  
Multi Objective ◽  
Evaluation Time ◽  
Surrogate Function

Abstract This paper investigates the integration of a surrogate-assisted multi-objective evolutionary algorithm (MOEA) and a parallel computation scheme to reduce the computing time until obtaining the optimal solutions in evolutionary algorithms (EAs). A surrogate-assisted MOEA solves multi-objective optimization problems while estimating the evaluation of solutions with a surrogate function. A surrogate function is produced by a machine learning model. This paper uses an extreme learning surrogate-assisted MOEA/D (ELMOEA/D), which utilizes one of the well-known MOEA algorithms, MOEA/D, and a machine learning technique, extreme learning machine (ELM). A parallelization of MOEA, on the other hand, evaluates solutions in parallel on multiple computing nodes to accelerate the optimization process. We consider a synchronous and an asynchronous parallel MOEA as a master-slave parallelization scheme for ELMOEA/D. We carry out an experiment with multi-objective optimization problems to compare the synchronous parallel ELMOEA/D with the asynchronous parallel ELMOEA/D. In the experiment, we simulate two settings of the evaluation time of solutions. One determines the evaluation time of solutions by the normal distribution with different variances. On the other hand, another evaluation time correlates to the objective function value. We compare the quality of solutions obtained by the parallel ELMOEA/D variants within a particular computing time. The experimental results show that the parallelization of ELMOEA/D significantly reduces the computational time. In addition, the integration of ELMOEA/D with the asynchronous parallelization scheme obtains higher quality of solutions quicker than the synchronous parallel ELMOEA/D.

scholarly journals Antitumor Trans Platinum Adducts of GMP and AMP

2000 ◽  
Vol 7 (4) ◽  
pp. 169-176 ◽  
Author(s):  
Yangzhong Liu ◽  
Maria F. Sivo ◽  
Giovanni Natile ◽  
Einar Sletten
Keyword(s):  
Antitumor Activity ◽  
Energy Minimization ◽  
Qualitative Agreement ◽  
Early Stage ◽  
2D Nmr ◽  
The Other ◽  
Coordination Geometry ◽  
Nmr Data ◽  
Monofunctional Adducts

Recently it has been shown that several analogues of the clinically ineffective trans-DDP exhibit antitumor activity comparable to that of cis-DDP. The present paper describes the binding of antitumor trans-[PtCl2(E-iminoether)2] (trans-EE) to guanosinemonophosphate (GMP) and adenosinemonophosphate (AMP). We have used HPLC and H1 and N15 NMR to characterize the different adducts. In the case of a 1:1 mixture of trans-EE and GMP, at an early stage of the reaction, a monofunctional adduct is formed which, subsequently, is partly converted into a monosolvated monofunctional species. After about 70 hours an equilibrium is established between chloro and solvato monofunctional adducts at a ratio of 30/70. In the presence of excess GMP (4:1) the initially formed monofunctional adducts react further to give two bifunctional adducts, one with the iminoether ligands in their original E configurations and the other with the iminoether ligands having one E and the other, Z configurations. The coordination geometry obtained by energy minimization calculations is in qualitative agreement with 2D NMR data.

Large-Strain Axisymmetric Deformation of Cylindrical Shells

1987 ◽  
Vol 54 (2) ◽  
pp. 287-291 ◽  
Author(s):  
G. W. Brodland ◽  
H. Cohen
Keyword(s):  
Nonlinear Equations ◽  
Energy Minimization ◽  
Parametric Analysis ◽  
Large Strain ◽  
Cylindrical Shells ◽  
Radial Force ◽  
Axisymmetric Deformation ◽  
The Other ◽  
Large Strains ◽  
Minimization Technique

Nonlinear equations are derived for the axisymmetric deformation of thin, cylindrical shells made of Mooney-Rivlin materials and subject to arbitrarily large strains and rotations. These equations are then implemented numerically using an energy minimization technique. Finally, an extensive parametric analysis is done of cylindrical shells which are clamped at one end and loaded with either a radial force or an edge moment uniformly distributed along the circumference of the other end.

Elimination of Branch, Grashof, and Order Defects in Path-Angle Generation and Function Generation Synthesis

1979 ◽  
Vol 101 (3) ◽  
pp. 428-437 ◽  
Author(s):  
K. J. Waldron ◽  
E. N. Stevensen
Keyword(s):  
Motion Generation ◽  
Function Generation ◽  
And Function ◽  
Generation Problem

Path-Angle Generation and Function Generation synthesis problems are restated as Plane-Position (or Motion Generation) problems, enabling the use of the classical Burmester technique and recent extensions that permit the avoidance of Branch, Grashof, and Order defects. An example of the solution of a Path-Angle Generation problem is given.

Sign in / Sign up

Export Citation Format

Share Document