A learning from demonstration framework for implementation of a feeding task

2018 ◽  
Vol 02 (01) ◽  
pp. 1850001 ◽  
Author(s):  
Nabil Ettehadi ◽  
Aman Behal
Keyword(s):  
Expectation Maximization ◽  
Optimization Problem ◽  
Gaussian Mixture ◽  
The Other ◽  
Learning From Demonstration ◽  
Constrained Optimization Problem ◽  
Gaussian Distributions ◽  
End Effector ◽  
Mixture Regression ◽  
Mixture Of Gaussian

In this paper, a learning from demonstration (LFD) approach is used to design an autonomous meal-assistance agent. The feeding task is modeled as a mixture of Gaussian distributions. Using the data collected via kinesthetic teaching, the parameters of the Gaussian mixture model (GMM) are learnt using Gaussian mixture regression (GMR) and expectation maximization (EM) algorithm. Reproduction of feeding trajectories for different environments is obtained by solving a constrained optimization problem. In this method we show that obstacles can be avoided by robot’s end-effector by adding a set of extra constraints to the optimization problem. Finally, the performance of the designed meal assistant is evaluated in two feeding scenario experiments: one considering obstacles in the path between the bowl and the mouth and the other without.

scholarly journals An Extended DMPs framework for Decoupled Quaternions Learning and generalization

2021 ◽  
Author(s):  
Zhiwei Liao ◽  
Fei Zhao ◽  
Gedong Jiang ◽  
Xuesong Mei
Keyword(s):  
Riemannian Manifolds ◽  
Gaussian Mixture ◽  
Joint Space ◽  
Robot Learning ◽  
Learning From Demonstration ◽  
End Effector ◽  
Mixture Regression ◽  
Cartesian Space ◽  
Dynamic Movement ◽  
Dynamic Movement Primitives

Abstract Dynamic Movement Primitives (DMPs) as a robust and efficient framework has been studied widely for robot learning from demonstration. Classical DMPs framework mainly focuses on the movement learning in Cartesian or joint space, and can't properly represent end-effector orientation. In this paper, we present an Extended DMPs framework (EDMPs) both in Cartesian space and Riemannian manifolds for Quaternion-based orientations learning and generalization. Gaussian Mixture Model and Gaussian Mixture Regression are adopted as the initialization phase of EDMPs to handle multi-demonstrations and obtain their mean and covariance. Additionally, some evaluation indicators including reachability and similarity are defined to characterize the learning and generalization abilities of EDMPs. Finally, the quaternion-based orientations are successfully transferred from human to the robot, and a real-world experiment is conducted to verify the effectiveness of the proposed method. The experimental results reveal that the presented approach can learn and generalize multi-space parameters under multi-demonstrations.

scholarly journals Statistical Forecasting : Third Wave of COVID-19-With an Application to India

2021 ◽  
Author(s):  
Sabara Parshad Rajeshbhai ◽  
Subhra Sankar Dhar ◽  
Shalabh Shalabh
Keyword(s):  
Gaussian Mixture Model ◽  
Gaussian Mixture ◽  
Third Wave ◽  
Statistical Methodology ◽  
Gaussian Distributions ◽  
Wave Modelling ◽  
The Third ◽  
Entire World ◽  
The Impact ◽  
Mixture Of Gaussian

The pandemic due to the SARS-CoV-2 virus impacted the entire world in different waves. An important question that arise after witnessing the first and second waves of COVID-19 is - Will the third wave also arrive and if yes, then when. Various types of methodologies are being used to explore the arrival of third wave. A statistical methodology based on the fitting of mixture of Gaussian distributions is explored in this paper and the aim is to forecast the third wave using the data on the first two waves of pandemic. Utilizing the data of different countries that are already facing the third wave, modelling of their daily cases data and predicting the impact and timeline for the third wave in India is attempted in this paper. The Gaussian mixture model based on algorithm for clustering is used to estimate the parameters.

Narrowband Emitter Identification by Exploiting Gaussian Mixture Model

2011 ◽  
Vol 65 ◽  
pp. 503-508
Author(s):  
Yu Yu Liao ◽  
Ke Xin Jia ◽  
Zi Shu He ◽  
Song Feng Deng
Keyword(s):  
Gaussian Mixture Model ◽  
Mixture Model ◽  
Expectation Maximization ◽  
Execution Time ◽  
Minimum Description Length ◽  
Gaussian Mixture ◽  
Frequency Information ◽  
Second Stage ◽  
Length Variant ◽  
Mixture Of Gaussian

Narrowband emitter identification is used to correctly identify unknown narrowband emitters from the results of direction finding (DF). In this paper, we modeled the set of azimuth angles by a mixture of Gaussian densities, and divided narrowband emitter identification into two different stages. In the first stage, a competitive stop expectation-maximization (CSEM) algorithm was developed, which was based on Shapiro-Wilk test and minimum description length variant (MDL2) criterion. The CSEM only employed the estimated azimuth angles at all the signal-occupied frequency bins as feature parameters, while the frequency information implied in each cluster was not exploited sufficiently. So based on the implied frequency information, a postprocessing algorithm was introduced in the second stage. The experimental results show that the CSEM algorithm is more robust, and it has an increased capability to find the underlying model while maintaining a low execution time. By adopting CSEM and postprocessing algorithm in narrowband emitter identification, we are able to identify narrowband emitters with high correctness.

scholarly journals Application of Lagrange Multiplier Method for Computing Fold Bifurcation Point in A Two-Prey One Predator Dynamical System

2018 ◽  
Vol 24 (2) ◽  
pp. 7-19
Author(s):  
Marwan Marwan ◽  
Johan Matheus Tuwankotta ◽  
Eric Harjanto
Keyword(s):  
Dynamical System ◽  
Lagrange Multiplier ◽  
Bifurcation Point ◽  
Optimization Problem ◽  
The Other ◽  
Lagrange Multiplier Method ◽  
Multiplier Method ◽  
Constrained Optimization Problem ◽  
Fold Bifurcation ◽  
The Cost

We propose by means of an example of applications of the classical Lagrange Multiplier Method for computing fold bifurcation point of an equilibrium ina one-parameter family of dynamical systems. We have used the fact that an equilibrium of a system, geometrically can be seen as an intersection between nullcline manifolds of the system. Thus, we can view the problem of two collapsing equilibria as a constrained optimization problem, where one of the nullclines acts as the cost function while the other nullclines act as the constraints.

scholarly journals Hybrid Metaheuristic for Designing an End Effector as a Constrained Optimization Problem

IEEE Access ◽  
2017 ◽  
Vol 5 ◽  
pp. 6002-6014 ◽  
Author(s):  
Eduardo Vega-Alvarado ◽  
Edgar Alfredo Portilla-Flores ◽  
Maria Barbara Calva-Yanez ◽  
Gabriel Sepulveda-Cervantes ◽  
Jorge Alexander Aponte-Rodriguez ◽  
...  
Keyword(s):  
Constrained Optimization ◽  
Optimization Problem ◽  
Constrained Optimization Problem ◽  
End Effector ◽  
Hybrid Metaheuristic

A new algorithm for finding CRM-model coefficients

2019 ◽  
Vol 5 (3) ◽  
pp. 164-185 ◽  
Author(s):  
Alexander D. Bekman ◽  
Sergey V. Stepanov ◽  
Alexander A. Ruchkin ◽  
Dmitry V. Zelenin
Keyword(s):  
Approximate Solution ◽  
Optimization Problem ◽  
Optimal Solution ◽  
Complex Form ◽  
Approximate Solutions ◽  
The Other ◽  
Programming Algorithm ◽  
Stochastic Algorithms ◽  
Large Set ◽  
Flow Rates

The quantitative evaluation of producer and injector well interference based on well operation data (profiles of flow rates/injectivities and bottomhole/reservoir pressures) with the help of CRM (Capacitance-Resistive Models) is an optimization problem with large set of variables and constraints. The analytical solution cannot be found because of the complex form of the objective function for this problem. Attempts to find the solution with stochastic algorithms take unacceptable time and the result may be far from the optimal solution. Besides, the use of universal (commercial) optimizers hides the details of step by step solution from the user, for example&nbsp;— the ambiguity of the solution as the result of data inaccuracy.<br> The present article concerns two variants of CRM problem. The authors present a new algorithm of solving the problems with the help of “General Quadratic Programming Algorithm”. The main advantage of the new algorithm is the greater performance in comparison with the other known algorithms. Its other advantage is the possibility of an ambiguity analysis. This article studies the conditions which guarantee that the first variant of problem has a unique solution, which can be found with the presented algorithm. Another algorithm for finding the approximate solution for the second variant of the problem is also considered. The method of visualization of approximate solutions set is presented. The results of experiments comparing the new algorithm with some previously known are given.

scholarly journals Imitation learning-based framework for learning 6-D linear compliant motions

2021 ◽  
Author(s):  
Markku Suomalainen ◽  
Fares J. Abu-dakka ◽  
Ville Kyrki
Keyword(s):  
Contact Force ◽  
Free Space ◽  
Prior Information ◽  
Goal Area ◽  
Imitation Learning ◽  
Learning From Demonstration ◽  
End Effector ◽  
Localization Accuracy ◽  
Novel Method ◽  
Contact Tasks

AbstractWe present a novel method for learning from demonstration 6-D tasks that can be modeled as a sequence of linear motions and compliances. The focus of this paper is the learning of a single linear primitive, many of which can be sequenced to perform more complex tasks. The presented method learns from demonstrations how to take advantage of mechanical gradients in in-contact tasks, such as assembly, both for translations and rotations, without any prior information. The method assumes there exists a desired linear direction in 6-D which, if followed by the manipulator, leads the robot’s end-effector to the goal area shown in the demonstration, either in free space or by leveraging contact through compliance. First, demonstrations are gathered where the teacher explicitly shows the robot how the mechanical gradients can be used as guidance towards the goal. From the demonstrations, a set of directions is computed which would result in the observed motion at each timestep during a demonstration of a single primitive. By observing which direction is included in all these sets, we find a single desired direction which can reproduce the demonstrated motion. Finding the number of compliant axes and their directions in both rotation and translation is based on the assumption that in the presence of a desired direction of motion, all other observed motion is caused by the contact force of the environment, signalling the need for compliance. We evaluate the method on a KUKA LWR4+ robot with test setups imitating typical tasks where a human would use compliance to cope with positional uncertainty. Results show that the method can successfully learn and reproduce compliant motions by taking advantage of the geometry of the task, therefore reducing the need for localization accuracy.

scholarly journals Nonconvex constrained optimization by a filtering branch and bound

2020 ◽  
Author(s):  
Gabriele Eichfelder ◽  
Kathrin Klamroth ◽  
Julia Niebling
Keyword(s):  
Constrained Optimization ◽  
Optimization Problem ◽  
Feasible Solution ◽  
Constrained Optimization Problem ◽  
Numerical Tests ◽  
Nonconvex Constrained Optimization ◽  
Objective Value ◽  
Feasible Solutions ◽  
Quality Guarantee ◽  
Single Objective

AbstractA major difficulty in optimization with nonconvex constraints is to find feasible solutions. As simple examples show, the $$\alpha $$ α BB-algorithm for single-objective optimization may fail to compute feasible solutions even though this algorithm is a popular method in global optimization. In this work, we introduce a filtering approach motivated by a multiobjective reformulation of the constrained optimization problem. Moreover, the multiobjective reformulation enables to identify the trade-off between constraint satisfaction and objective value which is also reflected in the quality guarantee. Numerical tests validate that we indeed can find feasible and often optimal solutions where the classical single-objective $$\alpha $$ α BB method fails, i.e., it terminates without ever finding a feasible solution.

Sign in / Sign up

Export Citation Format

Share Document