Structural and Material Optimization for Automatic Synthesis of Spine-Segment Mechanisms for Humanoid Robots with Custom Stiffness Profiles

Typical artificial joints for humanoid robots use actual human body joints only as an inspiration. The load responses of these structures rarely match those of the corresponding joints, which is important when applying the robots in environments tailored to humans. In this study, we proposed a novel, automated method for designing substitutes for a human intervertebral joint. The substitutes were considered as two platforms, connected by a set of flexible links. Their structural and material parameters were obtained through optimization with a structured Genetic Algorithm, based on the reference angular stiffnesses. The proposed approach was tested in three numerical scenarios. In the first test, a mechanism with angular stiffnesses corresponded to the actual L4–L5 intervertebral joint. Scenarios 2 and 3 featured mechanisms with geometry and structure comparable to the joint, but with custom stiffness profiles. The obtained results proved the effectiveness of the proposed method. It could be employed in the design of artificial joints for humanoid robots and orthotic structures for the human spine. As the approach is general, it could also be extended to different body joints.

Download Full-text

Symmetry Based Material Optimization

Symmetry ◽

10.3390/sym13020315 ◽

2021 ◽

Vol 13 (2) ◽

pp. 315

Author(s):

Zeyun Shi ◽

Jinkeng Lin ◽

Jiong Chen ◽

Yao Jin ◽

Jin Huang

Keyword(s):

Force Field ◽

Vector Fields ◽

General Situation ◽

Kernel Space ◽

Dynamic Simulations ◽

Physical Behavior ◽

Material Parameters ◽

Material Optimization ◽

The Asymmetry ◽

Asymmetric Mesh

Many man-made or natural objects are composed of symmetric parts and possess symmetric physical behavior. Although its shape can exactly follow a symmetry in the designing or modeling stage, its discretized mesh in the analysis stage may be asymmetric because generating a mesh exactly following the symmetry is usually costly. As a consequence, the expected symmetric physical behavior may not be faithfully reproduced due to the asymmetry of the mesh. To solve this problem, we propose to optimize the material parameters of the mesh for static and kinematic symmetry behavior. Specifically, under the situation of static equilibrium, Young’s modulus is properly scaled so that a symmetric force field leads to symmetric displacement. For kinematics, the mass is optimized to reproduce symmetric acceleration under a symmetric force field. To efficiently measure the deviation from symmetry, we formulate a linear operator whose kernel contains all the symmetric vector fields, which helps to characterize the asymmetry error via a simple ℓ2 norm. To make the resulting material suitable for the general situation, the symmetric training force fields are derived from modal analysis in the above kernel space. Results show that our optimized material significantly reduces the asymmetric error on an asymmetric mesh in both static and dynamic simulations.

Download Full-text

Analysis of gabor filter based features with PCA and GA for the detection of drusen in fundus images

International Journal of Engineering & Technology ◽

10.14419/ijet.v7i1.8969 ◽

2018 ◽

Vol 7 (1) ◽

pp. 115

Author(s):

Sheela N. ◽

Basavaraj L.

Keyword(s):

Genetic Algorithm ◽

Nearest Neighbor ◽

Gabor Filter ◽

Age Related Macular Degeneration ◽

Misclassification Rate ◽

Support Vector ◽

K Nearest Neighbor ◽

Automated Method ◽

Age Related ◽

Predictive Rate

Human eye can be affected by different types of diseases. Age-Related Macular Degeneration (AMD) is one of the such diseases, and it mainly occurs after 50 years of age. This disease is characterized by the occurrence of yellow spots called as Drusen. In this work, an automated method for the detection of drusen in Fundus image has been developed, and it has been tested on 70 images consisting of 30 normal images and 40 images with drusen. Performance of the Support Vector Machine (SVM) and K Nearest Neighbor (KNN) classifier has been evaluated using Data's reduction using Principle Component Analysis (PCA) and Data's selection using Genetic Algorithm (GA).Performance evaluation has been done in terms of accuracy, sensitivity, specificity, misclassification rate, positive predictive rate, negative predictive rate and Youden’s Index. The proposed method has achieved highest accuracy of 98.7% when data selection using Genetic Algorithm has been applied.

Download Full-text

Back Analysis of Material Parameters of Concrete Dam Using Genetic Algorithm

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.562-564.1955 ◽

2012 ◽

Vol 562-564 ◽

pp. 1955-1958

Author(s):

Jin Bao Liu ◽

Shou Ju Li ◽

Wei Zhu

Keyword(s):

Genetic Algorithm ◽

Objective Function ◽

Optimization Problem ◽

Back Analysis ◽

Water Level Change ◽

Concrete Dam ◽

Material Parameters ◽

Level Change ◽

Inversion Procedure ◽

The Difference

The inverse problem of parameter identification is deal with by minimizing an objective function that contains the difference between observed and calculated dam displacements. The optimization problem of minimizing objective function is solved with genetic algorithm. The calculated dam displacements are simulated by using finite element method according to water level change acting on dam upstream. The practical dam displacements are observed on the dam crest. The investigation shows that the forecasted dam displacements agree well with observed ones. The effectiveness of proposed inversion procedure is validated.

Download Full-text

Parameter Identification of Thermal Visco-Plastic Model Considering Dynamic Recrystallization

Materials Science Forum ◽

10.4028/www.scientific.net/msf.561-565.1869 ◽

2007 ◽

Vol 561-565 ◽

pp. 1869-1874

Author(s):

Quan Lin Jin ◽

Yan Shu Zhang

Keyword(s):

Numerical Simulation ◽

Genetic Algorithm ◽

Parameter Identification ◽

Optimization Methods ◽

Optimization Method ◽

Special Program ◽

Global Optimization Method ◽

Material Parameters ◽

Real Coded Genetic Algorithm ◽

Deformation Test

A hybrid global optimization method combining the Real-coded genetic algorithm and some classical local optimization methods is constructed and applied to develop a special program for parameter identification. Finally, the parameter identification for both 26Cr2Ni4MoV steel and AZ31D magnesium alloy is carried out by using the program. A comparison of deformation test and numerical simulation shows that the parameter identification and the obtained two sets of material parameters are all available.

Download Full-text

Synthesis of Bistable Periodic Structures Using Topology Optimization and a Genetic Algorithm

Volume 2: 31st Design Automation Conference, Parts A and B ◽

10.1115/detc2005-84904 ◽

2005 ◽

Author(s):

Jitendra Prasad ◽

Alejandro Diaz

Keyword(s):

Genetic Algorithm ◽

Topology Optimization ◽

Periodic Structures ◽

Performance Criterion ◽

Ground Structure ◽

Structure Approach ◽

Automatic Synthesis ◽

Compliant Structure ◽

Beam Elements ◽

Ground Structure Approach

Formulations for the automatic synthesis of two-dimensional bistable, compliant periodic structures are presented, based on standard methods for topology optimization. The design space is parameterized using non-linear beam elements and a ground structure approach. A performance criterion is suggested, based on characteristics of the load-deformation curve of the compliant structure. A genetic algorithm is used to find candidate solutions. A numerical implementation of this methodology is discussed and illustrated using a simple example.

Download Full-text

Optimization of cooperative motion for humanoid robots using a genetic algorithm

10.1117/12.664773 ◽

2005 ◽

Cited By ~ 1

Author(s):

Kook-Jin Choi ◽

Yoon-Kwon Hwang ◽

Dae Sun Hong ◽

Won Jee Chung ◽

Il-Hwan Park

Keyword(s):

Genetic Algorithm ◽

Humanoid Robots ◽

Cooperative Motion

Download Full-text

Optimized material management in construction using multi-layer perceptron

Canadian Journal of Civil Engineering ◽

10.1139/cjce-2018-0149 ◽

2019 ◽

Vol 46 (10) ◽

pp. 909-923

Author(s):

Farzaneh Golkhoo ◽

Osama Moselhi

Keyword(s):

Genetic Algorithm ◽

Construction Material ◽

Construction Materials ◽

Multi Layer Perceptron ◽

Material Management ◽

Control Material ◽

Automated Method ◽

Purchase Cost ◽

Trade Offs ◽

Material Management System

Construction material represents a major component of the project cost. Therefore, it is essential to control material on construction job sites. Efficient material management system requires trade-offs and optimized balance among elements of material cost including purchase cost, storage cost, opportunity cost, ordering cost, and unavailability cost. Thus, there is a need to develop an automated method for optimizing the delivery and inventory of construction materials not only in the planning phase but also in the construction phase to account for introduced changes. In this research a novel genetic algorithm – multi-layer perceptron (GA-MLP) method is proposed to generate optimized material delivery schedule. Multi-layer perceptron (MLP) is utilized to improve genetic algorithm (GA) by generating memory to overcome local minima encountered in applying GA for optimization. This automated method supports contractors to buy construction materials with the least cost and without leading to material shortage or surplus. The proposed automated method has been validated through a numerical example. The obtained results demonstrate that GA-MLP outperform GA in optimizing construction material inventory.

Download Full-text