Filters








624 Hits in 5.3 sec

Can a Compact Neuronal Circuit Policy be Re-purposed to Learn Simple Robotic Control? [article]

Ramin Hasani, Mathias Lechner, Alexander Amini, Daniela Rus, Radu Grosu
2019 arXiv   pre-print
Inspired by the structure of the nervous system of the soil-worm, C. elegans, we introduce Neuronal Circuit Policies (NCPs), defined as the model of biological neural circuits reparameterized for the control  ...  We learn instances of NCPs to control a series of robotic tasks, including the autonomous parking of a real-world rover robot.  ...  INTRODUCTION We wish to explore a new class of machine learning algorithms for robot control that is inspired by nature.  ... 
arXiv:1809.04423v2 fatcat:hefpnk2nmnderp73rgmvh2c524

Mobile Robot Navigation and Obstacle Avoidance Techniques: A Review

Anish Pandey
2017 International Robotics & Automation Journal  
Mobile robot is an autonomous agent capable of navigating intelligently anywhere using sensor-actuator control techniques.  ...  Several techniques have been applied by the various researchers for mobile robot navigation and obstacle avoidance.  ...  They have used the fuzzy rule-based controller to interpret sensory information, and neural network controls the heading angle of the robot during navigation. Baturone et al.  ... 
doi:10.15406/iratj.2017.02.00023 fatcat:m6viumq36zf5zbfeexua475gjy

Collective control of modular soft robots via embodied Spiking Neural Cellular Automata [article]

Giorgia Nadizar, Eric Medvet, Stefano Nichele, Sidney Pontes-Filho
2022 arXiv   pre-print
In this work, we propose a novel form of collective control, influenced by Neural Cellular Automata (NCA) and based on the bio-inspired Spiking Neural Networks: the embodied Spiking NCA (SNCA).  ...  Voxel-based Soft Robots (VSRs) are a form of modular soft robots, composed of several deformable cubes, i.e., voxels.  ...  SPIKING NEURAL NETWORKS AS ROBOTIC CONTROLLERS Spiking Neural Networks (SNNs) are a type of ANNs in which biological resemblance plays a fundamental role (Gerstner & Kistler, 2002) .  ... 
arXiv:2204.02099v1 fatcat:jlk45g2lkrgebhwiy2vleopbqq

A Natural Lottery Ticket Winner: Reinforcement Learning with Ordinary Neural Circuits

Ramin M. Hasani, Mathias Lechner, Alexander Amini, Daniela Rus, Radu Grosu
2020 International Conference on Machine Learning  
ONC networks are compact, 77% sparser than their counterpart neural controllers, and their neural dynamics are fully interpretable at the cell-level.  ...  Inspired by the structure of the nervous system of the soilworm, C. elegans, we introduce ordinary neural circuits (ONCs), defined as the model of biological neural circuits reparameterized for the control  ...  Introduction We wish to explore a new class of machine learning algorithms for robot control inspired by nature.  ... 
dblp:conf/icml/HasaniLARG20 fatcat:gvx3wg5bvbbvlkbu5znqqlyqyi

Criticality-Driven Evolution of Adaptable Morphologies of Voxel-Based Soft-Robots

Jacopo Talamini, Eric Medvet, Stefano Nichele
2021 Frontiers in Robotics and AI  
networks.  ...  The paradigm of voxel-based soft robots has allowed to shift the complexity from the control algorithm to the robot morphology itself.  ...  Neural network sensing controllers allow the robots with TABLE 3 | Results for the optimized and randomly generated morphologies coupled with the neural controller.  ... 
doi:10.3389/frobt.2021.673156 fatcat:fhbfrmqmsnexdop2ajqeaxrdjy

Collective Intelligence for Deep Learning: A Survey of Recent Developments [article]

David Ha, Yujin Tang
2022 arXiv   pre-print
However, as these neural networks become bigger, more complex, and more widely used, fundamental problems with current deep learning models become more apparent.  ...  Advances in artificial neural networks alongside corresponding advances in hardware accelerators with large memory capacity, together with the availability of large datasets enabled practitioners to train  ...  Wang et al. 91 and Huang et al. 32 explored the use of modular neural networks to control each individual actuator of a simulated robot for continuous control.  ... 
arXiv:2111.14377v3 fatcat:dg5uvn7mt5g5ncgtrzw3a3ul4y

On exploration of geometrically constrained space by medicinal leeches Hirudo verbana

Andrew Adamatzky
2015 Biosystems (Amsterdam. Print)  
Leeches could be ideal blue-prints for designing flexible soft robots which are modular, multi-functional, fault-tolerant, easy to control, capable for navigating using optical, mechanical and chemical  ...  With future designs of leech-robots in mind we study how leeches behave in geometrically constrained spaces.  ...  Representative examples include climbing worm robots controlled by oscillatory networks [67] , worm robots propagating in flexible environments [82] , neumatic flexible robot prototype for pipes inspection  ... 
doi:10.1016/j.biosystems.2015.02.005 pmid:25766395 fatcat:hcxrcyf25zcnzonu2q4hbe6zly

Robots in invertebrate neuroscience

Barbara Webb
2002 Nature  
Robots can be used to test hypotheses in the neuroscience of sensorimotor control. Some well explored systems in invertebrates are particularly suited to such implementations.  ...  complementary insight into understanding these complex systems, by providing a real-world grounding and thus emphasising the contribution of the physics of environments, sensors and actuators to the control  ...  The neural network used to control the behaviour is a simplification of identified neural properties in the nematode.  ... 
doi:10.1038/417359a pmid:12015617 fatcat:x7pp45ngnfdnjli524pf4qt4me

Spiking neural state machine for gait frequency entrainment in a flexible modular robot [article]

Alex Spaeth, Maryam Tebyani, David Haussler, Mircea Teodorescu
2020 arXiv   pre-print
We propose a modular architecture for neuromorphic closed-loop control based on bistable relaxation oscillator modules consisting of three spiking neurons each.  ...  By combining these modules, we can construct a neural state machine capable of generating the cyclic or repetitive behaviors necessary for legged locomotion.  ...  The position of the robot over time during this experiment is shown in Fig 11. Conclusion We have described a simple approach to the design of spiking neural networks for robotic control.  ... 
arXiv:2007.07346v2 fatcat:hdqf4igla5hapcgbydkhtoenp4

Spiking neural state machine for gait frequency entrainment in a flexible modular robot

Alex Spaeth, Maryam Tebyani, David Haussler, Mircea Teodorescu, Gennady Cymbalyuk
2020 PLoS ONE  
We propose a modular architecture for neuromorphic closed-loop control based on bistable relaxation oscillator modules consisting of three spiking neurons each.  ...  By combining these modules, we can construct a neural state machine capable of generating the cyclic or repetitive behaviors necessary for legged locomotion.  ...  The position of the robot over time during this experiment is shown in Fig 11. Conclusion We have described a simple approach to the design of spiking neural networks for robotic control.  ... 
doi:10.1371/journal.pone.0240267 pmid:33085673 fatcat:4c7maqdvznafzksxkla4b6lcfm

Sensorimotor transformations in the worlds of frogs and robots

Michael A. Arbib, Jim-Shih Liaw
1995 Artificial Intelligence  
We present a set of biological design principles within a broader perspective that shows their relevance for robot design.  ...  An explicit account of neural mechanism of avoidance behavior in the frog illustrates how schemas may be implemented in neural networks.  ...  the approach to the design of robot controllers offered by Brooks [9] .  ... 
doi:10.1016/0004-3702(94)00055-6 fatcat:tnuagftutfg75hdoo3hxwjr7cu

Large-scale, Small-scale Systems [chapter]

Jim Austin, Dave Cliff, Robert Ghanea-Hercock, Andy Wright
2006 Cognitive Systems - Information Processing Meets Brain Science  
Acknowledgements The authors thank Flaviu Adrian Marginean for his help and background work in the construction of this document.  ...  an artificial neural network.  ...  We can trace the demonstration of minimal biologically inspired architectures for mobile robot controllers back to the cybernetics research of Walter's (1950 Walter's ( , 1951 'turtle' robots and Ashby's  ... 
doi:10.1016/b978-012088566-4/50005-2 fatcat:mr7mjjmsingjxoiia3yrblsvbq

Energy Optimization in Wireless Sensor Network through Natural Science Computing: A Survey

Sanjeev Wagh, Ramjee Prasad
2013 Journal of green engineering  
It integrates contrasting techniques of Genetic systems, Neural System, Immune systems and Cellular automata inspired from human organs to give solutions for energy conservation problems in Wireless sensor  ...  Nature-inspired computational algorithms are attracting engineering research for artificial intelligence solutions.  ...  Conclusions The paper has introduced the nature science inspired methods for controlling and managing wireless sensor networks.  ... 
doi:10.13052/jge1904-4720.342 fatcat:yqfecenhkzcnlboif3ueh7torq

KI 2007 Ankündigung

Joachim Hertzberg
2007 Künstliche Intelligenz  
But it also comprises biological inspirations e.g. for robot design, artificial neural networks, or emergent intelligence, as well as logical underpinnings of automated deduction and knowledge representation  ...  Neural Networks and Bayesian Networks are used for static code analysis in order to determine whether a file contain malicious code.  ...  The goal is to provide a forum for the presentation of research as well as of existing and future applications and for lively discussions among researchers and industry.  ... 
dblp:journals/ki/Hertzberg07a fatcat:p5edhdrexjd4bh3aetqmymav2e

Intelligent control of a highly flexible robotic structure with hundreds of motor elements

Selahattin Ozcelik, Michael Blackburn, Grant R. Gerhart, Charles M. Shoemaker, Douglas W. Gage
2005 Unmanned Ground Vehicle Technology VII  
We then suggest two methods of intelligent control to manage the many motor elements. One method derives from neural networks, the other involves algorithms inspired by the biological immune system.  ...  This new robotic architecture possesses a variably compliant structure that allows for the controlled distribution of loads and forces, and for the maintenance of different conformations.  ...  Neural Network Controller A neural network controller is ideal for this type of distributed control problem in which the transfer functions must adapt on multiple time scales.  ... 
doi:10.1117/12.602608 fatcat:vijxhq7cgvenlbisztsywlh7ze
« Previous Showing results 1 — 15 out of 624 results