2,364 Hits in 3.1 sec

Semi-parametric Gaussian process for robot system identification

Tingfan Wu, Javier Movellan
2012 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems  
Here we present an approach to robot system identification, named Semi-Parametric Gaussian Processes (SGP), that elegantly combines the advantages of parametric and nonparametric approaches.  ...  Recent nonparametric machine learning approaches to system identification have shown good promise, outperforming parameterized mathematical models when applied to complex robot system identification problems  ...  Gaussian Process has been used widely for system identification in robotics.  ... 
doi:10.1109/iros.2012.6385977 dblp:conf/iros/WuM12 fatcat:2m26mmiyunfgtd72motfifidcm

Cascaded Gaussian Processes for Data-efficient Robot Dynamics Learning [article]

Sahand Rezaei-Shoshtari, David Meger, Inna Sharf
2019 arXiv   pre-print
Motivated by the recursive Newton-Euler formulation, we propose a novel cascaded Gaussian process learning framework for the inverse dynamics of robot manipulators.  ...  The learned modeling is tested in conjunction with the classical inverse dynamics model (semi-parametric) and on its own (non-parametric) in the context of feed-forward control of the arm.  ...  ACKNOWLEDGMENTS This work was supported by the National Sciences and Engineering Research Council (NSERC) Canadian Robotics Network (NCRN).  ... 
arXiv:1910.02291v1 fatcat:idfmp74ubjfcxdx56di3ntnbre

Online semi-parametric learning for inverse dynamics modeling [article]

Diego Romeres and Mattia Zorzi and Raffaello Camoriano and Alessandro Chiuso
2016 arXiv   pre-print
This paper presents a semi-parametric algorithm for online learning of a robot inverse dynamics model. It combines the strength of the parametric and non-parametric modeling.  ...  We provide an extensive comparison with other methods from the literature using real data from the iCub humanoid robot.  ...  The authors gratefully acknowledge the iCub Facility and LCSL-IIT@MIT research groups led by Francesco Nori and Lorenzo Rosasco for making their data and code available to us.  ... 
arXiv:1603.05412v2 fatcat:qjpw22mis5am3ktvfawmxqy67q

Autonomous Precision Pouring from Unknown Containers

Monroe Kennedy, Karl Schmeckpeper, Dinesh Thakur, Chenfanfu Jiang, Vijay Kumar, Kostas Daniilidis
2019 IEEE Robotics and Automation Letters  
This system is implemented on the Rethink Robotics Sawyer and KUKA LBR iiwa manipulators. Index Terms-Model learning for control, motion control, manipulation planning, service robots.  ...  Three methods are compared for estimating the volume-angle profile, and it is shown that a combination of online system identification and leveraged model priors results in reliable performance.  ...  We show that by combining online system identification and model priors through a Gaussian process, we can maximize performance with the specified system without tuning parameters for a given container  ... 
doi:10.1109/lra.2019.2902075 fatcat:t56sl2hb6jcqljjub7ahstmww4

Pseudospectral Model Predictive Control under Partially Learned Dynamics [article]

Manan Gandhi, Yunpeng Pan, Evangelos Theodorou
2017 arXiv   pre-print
In order to manage parametric uncertainty, we introduce an algorithm that utilizes Sparse Spectrum Gaussian Processes (SSGP) for online learning after each rollout.  ...  This paper aims to take the middle ground between these two approaches by introducing a semi-parametric representation of the underlying system dynamics.  ...  In these frameworks, the measurements taken during system identification and control tasks are used to build a dynamical model for the system.  ... 
arXiv:1702.04800v1 fatcat:6lhnn6nghbdqlalmxwx3uq3lye

Nonlinear gray-box identification using local models applied to industrial robots

Erik Wernholt, Stig Moberg
2011 Automatica  
We therefore propose an identification procedure that uses intermediate local models that allow for data compression and a less complex optimization problem.  ...  The weighted logarithmic least squares estimator achieves the best result and the identified model gives a good global description of the dynamics in the frequency range of interest for robot control.  ...  The main motivation for our study is identification of accurate dynamic models for industrial robots, which incorporate all the four mentioned properties.  ... 
doi:10.1016/j.automatica.2011.01.021 fatcat:2fzkpfkvj5b63mx6aygkxpcsxe

Identification and Stochastic Optimizing the UAV Motion Control in Turbulent Atmosphere

Yevgeny Somov, Nikolay Rodnishchev, Tatyana Somova
2021 International Journal of Aviation Science and Technology  
In a class of diffusion Markov processes, we formulate a problem of identification of nonlinear stochastic dynamic systems with random parameters, multiplicative and additive noises, control functions,  ...  The developed engineering methods for identification and optimizing nonlinear stochastic systems are presented as well as their application for unmanned aerial vehicles under wind disturbances caused by  ...  ., 2017, Identification of Elaborated methods for identification of the parameters Parameters and Control for Stochastic Dynamical and control functions of nonlinear stochastic systems  ... 
doi:10.23890/ijast.vm02is02.0202 fatcat:ykbkkbabzjcedeiwbcsp6kfq5i

Nonparametric Bayesian Method for Robot Anomaly Diagnose [chapter]

Xuefeng Zhou, Hongmin Wu, Juan Rojas, Zhihao Xu, Shuai Li
2020 Nonparametric Bayesian Learning for Collaborative Robot Multimodal Introspection  
Zhou et al., Nonparametric Bayesian Learning for Collaborative Robot Multimodal Introspection, https://doi.  ...  We evaluated the proposed methods with a multi-step human-robot collaboration objects kitting task on Baxter robot, the performance and results are presented of each method respectively. © The Author(s  ...  In robotics, complex semi-autonomous systems that have provided extensive assistance to humans, anomalies are occur occasionally [39] .  ... 
doi:10.1007/978-981-15-6263-1_5 fatcat:gfvoox3mifhc3dutbcxastriu4

Interactive perception based on Gaussian Process classification for house-hold objects recognition & sorting

Aamir Khan, Li Sun, Gerardo Aragon-Camarasa, J. Paul Siebert
2016 2016 IEEE International Conference on Robotics and Biomimetics (ROBIO)  
We present an interactive perception model for object sorting based on Gaussian Process (GP) classification that is capable of recognizing objects categories from point cloud data.  ...  Multi-class Gaussian Process classification is employed to provide and probable estimation of the identity of the object and serves a key role in the interactive perception cycle -modelling perception  ...  From these semi-autonomous sorting experiments, the proposed Gaussian Process based interactive sorting system outperformed random sorting by upto 30% in terms of sorting accuracy.  ... 
doi:10.1109/robio.2016.7866470 dblp:conf/robio/KhanSAS16 fatcat:vkhu2yz2avcldlsd2chkuqeigi

Gaussian Process based Passivation of a Class of Nonlinear Systems with Unknown Dynamics [article]

Thomas Beckers, Sandra Hirche
2018 arXiv   pre-print
For this purpose, we use the highly flexible, data-driven Gaussian process regression for the identification of the unknown dynamics for feed-forward compensation.  ...  The closed loop system of the nonlinear system, the Gaussian process model and a feedback control law is guaranteed to be semi-passive with a specific probability.  ...  The identification of dynamical systems with Gaussian processes is performed in [12] but without considering stability or passivity.  ... 
arXiv:1811.06648v1 fatcat:r3tpxiwf55g75pfvmw6kk5jfa4

Gaussian Process based Passivation of a Class of Nonlinear Systems with Unknown Dynamics

Thomas Beckers, Sandra Hirche
2018 2018 European Control Conference (ECC)  
For this purpose, we use the highly flexible, datadriven Gaussian process regression for the identification of the unknown dynamics for feed-forward compensation.  ...  The closed loop system of the nonlinear system, the Gaussian process model and a feedback control law is guaranteed to be semipassive with a specific probability.  ...  The identification of dynamical systems with Gaussian processes is performed in [12] but without considering stability or passivity.  ... 
doi:10.23919/ecc.2018.8550311 dblp:conf/eucc/BeckersH18 fatcat:hoparhcii5glbib7syzspavfcm

Preface [chapter]

2021 Soft Computing in Smart Manufacturing  
It addresses the heterogeneous complementary aspects such as control, monitoring, and modeling of different manufacturing tasks, including intelligent robotic systems and processes, modeling and parametric  ...  The demand for product personalization introduces a high level of uncertainty and requires highly adaptive, agile processes and systems.  ...  Neural Networks (CNN), for advanced object identification.  ... 
doi:10.1515/9783110693225-202 fatcat:75g7a5fdcfd5pgtbzu2f5elpvu

Multi-Sparse Gaussian Process: Learning based Semi-Parametric Control [article]

Mouhyemen Khan, Akash Patel, Abhijit Chatterjee
2020 arXiv   pre-print
In this paper, we propose a semi-parametric framework exploiting sparsity for learning-based control.  ...  Multi-Sparse Gaussian Process (MSGP) divides the original dataset into multiple sparse models with unique hyperparameters for each model.  ...  Semi-parametric methods have been applied for inverse dynamics [4] , [15] , system identification [16] , and forward dynamics [17] ; all using standard GPs.  ... 
arXiv:2003.01802v1 fatcat:dixyk34r3jf47d4vhzipthn7sa

A Latent State-Based Multimodal Execution Monitor with Anomaly Detection and Classification for Robot Introspection

Hongmin Wu, Yisheng Guan, Juan Rojas
2019 Applied Sciences  
Robot introspection is expected to greatly aid longer-term autonomy of autonomous manipulation systems.  ...  Both implementations are derived from Bayesian non-parametric methods with strong modeling capabilities for learning and inference of multivariate time series with complex and uncertain behavior patterns  ...  Acknowledgments: We thank Shuangqi Luo and Shuangda Duan for their assistance throughout this project. Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/app9061072 fatcat:rcca4rjkgralbhmnbishzigo6y

On-line regression algorithms for learning mechanical models of robots: A survey

Olivier Sigaud, Camille Salaün, Vincent Padois
2011 Robotics and Autonomous Systems  
With the emergence of more challenging contexts for robotics, the mechanical design of robots is becoming more and more complex.  ...  In particular, we provide a unified view of all recent algorithms that outlines their distinctive features and provides a framework for their combination.  ...  Furthermore, their semi-parametric regression approach does not alleviate the methodological difficulty of performing identification, which, in most cases, is not straightforward.  ... 
doi:10.1016/j.robot.2011.07.006 fatcat:mi46xi5l7bep5ijfzbdgiv7dqa
« Previous Showing results 1 — 15 out of 2,364 results