47 Hits in 3.2 sec

Machine Learning Algorithms in Bipedal Robot Control

Shouyi Wang, Wanpracha Chaovalitwongse, Robert Babuska
2012 IEEE Transactions on Systems Man and Cybernetics Part C (Applications and Reviews)  
Guidelines for future research on learning control of bipedal robots are provided in the end.  ...  This paper gives a review of recent advances on the stateof-the-art learning algorithms and their applications to bipedal robot control.  ...  [112] trained a multilayer perceptron (MLP) to learn a predesigned controller for a 3-link bipedal robot via a standard backpropagation (BP) algorithm.  ... 
doi:10.1109/tsmcc.2012.2186565 fatcat:tchoesxg6rc2vkuh7gtlxxfosa


Dusko Katić, Miomir Vukobratović
2005 IFAC Proceedings Volumes  
algorithms) in the area of humanoid robotic systems.  ...  Overall, this survey covers a broad selection of examples that will serve to demonstrate the advantages and disadvantages of the application of intelligent control techniques.  ...  Various types of neural networks are used for gait synthesis and control design of humanoid robots such as multilayer perceptrons, CMAC (Cerebellar Model Arithmetic Controller) networks, recurrent neural  ... 
doi:10.3182/20050703-6-cz-1902.01276 fatcat:wrzes2vitbf4rkkfn3c6evxuxe

Refined Continuous Control of DDPG Actors via Parametrised Activation

Mohammed Hossny, Julie Iskander, Mohamed Attia, Khaled Saleh, Ahmed Abobakr
2021 AI  
Continuous action spaces impose a serious challenge for reinforcement learning agents.  ...  There was no apparent improvement in Pendulum-v0 environment but the proposed method produces a more stable actuation signal compared to the state-of-the-art method.  ...  Also, the actuation layer is simply regarded as the final activation function π A (x) = tanh(x), and thus, the actor is typically modelled as one multilayer perceptron neural network (MLP).  ... 
doi:10.3390/ai2040029 fatcat:6qghubejqrda3auagpvwj43fme

Hybrid Model for Passive Locomotion Control of a Biped Humanoid:The Artificial Neural Network Approach

Manish Raj, Vijay Bhaskar-Semwal, G.C. Nandi
2018 International Journal of Interactive Multimedia and Artificial Intelligence  
Developing a correct model for a biped robot locomotion is extremely challenging due to its inherently unstable structure because of the passive joint located at the unilateral foot-ground contact and  ...  The present research describes the development of a hybrid biped model using an Open Dynamics Engine (ODE) based analytical three link leg model as a base model and, on top of it, an Artificial Neural  ...  The perceptron algorithm has been used since a long time and has its roots in the 1950s. A multilayer perceptron is a feed forward neural system with one or more concealed layers.  ... 
doi:10.9781/ijimai.2017.10.001 fatcat:cvcivce7xzfbbndc4kvnpb33mi

Neuro-Fuzzy Algorithm For A Biped Robotic System

Hataitep Wongsuwarn, Djitt Laowattana
2008 Zenodo  
This paper summaries basic principles and concepts of intelligent controls, implemented in humanoid robotics as well as recent algorithms being devised for advanced control of humanoid robots.  ...  Subsequently, we determine a relationship between joint trajectories and located forces on robot-s foot through a proposed neuro-fuzzy technique.  ...  Bantoon Srisuwan, Paisarn Suwantep and Thitisak Chanprom for their investigation of FHR-1 and their pioneering designs in controller and hardware interfacing.  ... 
doi:10.5281/zenodo.1083428 fatcat:75zt5ew5bveiti5szcuigsivoa

Brain-machine interfaces: an overview

Mikhail Lebedev
2014 Translational Neuroscience  
BMIs have a broad range of clinical goals, as well as the goal to enhance normal brain functions.  ...  BMIs are also classified as noninvasive or invasive according to the degree of their interference with the biological tissue.  ...  Figure 2 illustrates the settings of this experiment. BMI for bipedal walking Rhesus monkeys were trained to walk bipedally on a treadmill.  ... 
doi:10.2478/s13380-014-0212-z fatcat:7gqv37xlzfbqtobakl2htxqxzi

CIS-RAM 2019 Front Matter

2019 2019 IEEE International Conference on Cybernetics and Intelligent Systems (CIS) and IEEE Conference on Robotics, Automation and Mechatronics (RAM)  
Lastly, we thank the efforts of the organizing committee, student helpers and anyone who contributed one way or another, for making IEEE CIS-RAM 2019 a success!  ...  We thank the professional contributions from the members of the international advisory committee, authors, reviewers, associate editors, chairs of the regular and invited sessions, for making the high-quality  ...  The operator learns a robot-specific inverse dynamics model to compensate for the nonlinearities of the robot, and simultaneously learns a feed back control component that is specific to the successful  ... 
doi:10.1109/cis-ram47153.2019.9095827 fatcat:f2tvaddmhra77fnnyxywhepg4q

BALLU2: A Safe and Affordable Buoyancy Assisted Biped

Hosik Chae, Min Sung Ahn, Donghun Noh, Hyunwoo Nam, Dennis Hong
2021 Frontiers in Robotics and AI  
The paper describes the nonconventional characteristics of BALLU as a legged robot and then gives an analysis of its unique behavior.  ...  BALLU is a robot that never falls down due to the buoyancy provided by a set of helium balloons attached to the lightweight body, which solves many issues that hinder current robots from operating close  ...  Both networks individually consist of a multilayer perceptron with ELU nonlinear activation functions and ADAM optimizer with MSE (Minimum Square Error) loss and a constant learning rate.  ... 
doi:10.3389/frobt.2021.730323 pmid:34957224 pmcid:PMC8692890 fatcat:42hnpy4pjjeejexfngpl6jieza

A Machine Learning Approach for Improving the Movement of Humanoid NAO's Gaits

Fatmah Abdulrahman Baothman, Deepak Gupta
2021 Wireless Communications and Mobile Computing  
NAO is a humanoid bipedal robot designed to participate in football competitions against humans by 2050, and speed is crucial for football sports.  ...  With 12 attributes, the maximum accuracy metric rate of 65.31% was reached with only four hidden layers in 500 training cycles with a 0.5 learning rate for the best walking learning process, and the ANN  ...  The project is funded by the National Plan for Science, Technology, and Innovation (MAARI-FAH), King Abdulaziz City for Science and Technology, the Kingdom of Saudi Arabia (award number (03-INF188-08))  ... 
doi:10.1155/2021/1496364 fatcat:qp6zh4jnu5etnfek5kl36p3awy

Learning Torque Control for Quadrupedal Locomotion [article]

Shuxiao Chen, Bike Zhang, Mark W. Mueller, Akshara Rai, Koushil Sreenath
2022 arXiv   pre-print
However, the low frequency of such a policy hinders the advancement of highly dynamic locomotion behaviors.  ...  The design of most learning-based locomotion controllers adopts the joint position-based paradigm, wherein a low-frequency RL policy outputs target joint positions that are then tracked by a high-frequency  ...  ACKNOWLEDGMENTS We would like to thank Professor Stuart Russell and Professor Pieter Abbeel for their enlightening feedback during the course of this project, and Ayush Agrawal for his help with the experimental  ... 
arXiv:2203.05194v1 fatcat:zqk7mmhhfvf2hcrueskhrse2je

Robust Biped Locomotion Using Deep Reinforcement Learning on Top of an Analytical Control Approach [article]

Mohammadreza Kasaei, Miguel Abreu, Nuno Lau, Artur Pereira, Luis Paulo Reis
2021 arXiv   pre-print
The core of this framework is a specific dynamics model which abstracts a humanoid's dynamics model into two masses for modeling upper and lower body.  ...  Furthermore, a learning framework is developed based on Genetic Algorithm (GA) and Proximal Policy Optimization (PPO) to find the optimum parameters and to learn how to improve the stability of the robot  ...  The policy is represented by a multilayer perceptron with two hidden layers of 64 neurons.  ... 
arXiv:2104.10592v1 fatcat:ulkxmgotivdsrjdnysqimer7b4

Brain-computer interface: the future in the present
Интерфейс мозг–компьютер: будущее в настоящем

O. S. Levitskaya, M. A. Lebedev
2016 Bulletin of Russian State Medical University  
Brain-computer interfaces (BCIs) are a promising technology intended for the treatment of diseases and trauma of the nervous system.  ...  BCIs of various kinds are currently developing very rapidly, aided by the progress in computer science, robotic applications, neurophysiological techniques for recording brain activity and mathematical  ...  neighbour algorithm, artificial neural networks, multilayer perceptron, elements of fuzzy logic.  ... 
doi:10.24075/brsmu.2016-02-01 fatcat:arssbfnrnzewjor5xodbpiwds4

2008 Index IEEE Transactions on Automatic Control Vol. 53

2008 IEEE Transactions on Automatic Control  
The Author Index contains the primary entry for each item, listed under the first author's name.  ...  The primary entry includes the coauthors' names, the title of the paper or other item, and its location, specified by the publication abbreviation, year, month, and inclusive pagination.  ...  ., +, TAC Oct. 2008 1998-2011 Multilayer perceptrons Asymptotic Tracking for Uncertain Dynamic Systems Via a Multilayer Neural Network Feedforward and RISE Feedback Control Structure. Patre, P.  ... 
doi:10.1109/tac.2008.2010902 fatcat:5d76tgkjnvhohntuilcvl3jb3i

Neural network ARMAX model for a Furuta pendulum

David Acosta Villamil, Jovanny Pacheco Bolivar, Jose Noguera Polania, Marco Sanjuan Mejia
2021 Ingeniare : Revista Chilena de Ingeniería  
The authors present a model for the Furuta Pendulum using the equations of Euler-Lagrange and the methodology to identify a black-box model by training an NNARMAX (Neural Network Auto-Regressive Moving  ...  Due to its nonlinear nature, open-loop instability, and because it is an under-actuated system (more degrees of freedom than actuators), which is the basis for the design of vehicles such as the Segway  ...  , in motorcycle systems stabilization [15] , in bipedal robot systems [16] , in the famous conveyance known like segway [17, 18, 19] and [20] , this is a two wheels self-balancing system.  ... 
doi:10.4067/s0718-33052021000400668 fatcat:545tnjesf5hkbfsbczdpzflune

OptLayer - Practical Constrained Optimization for Deep Reinforcement Learning in the Real World [article]

Tu-Hoa Pham, Giovanni De Magistris, Ryuki Tachibana
2018 arXiv   pre-print
We demonstrate the effectiveness of our approach on robot reaching tasks, both simulated and in the real world.  ...  motions, since unconstrained trial-and-error interactions in the real world can have undesirable consequences for the robot or its environment.  ...  Throughout this paper, we take as neural network N a simple multilayer perceptron (MLP) with two hidden layers of size 32 each.  ... 
arXiv:1709.07643v2 fatcat:atnisgayxfco3gfsyw2uynkywa
« Previous Showing results 1 — 15 out of 47 results