6,398 Hits in 4.2 sec

Memetic robot control evolution and adaption to reality

Else-Line Ruud, Eivind Samuelsen, Kyrre Glette
2016 2016 IEEE Symposium Series on Computational Intelligence (SSCI)  
On the direct results from evolution in simulation, Lamarckian learning showed promising results, with a significant increase in final fitness compared with the results from evolution without learning.  ...  This paper demonstrates that Lamarckian evolution is effective in improving the performance of robot controller evolution, and that the same learning process on the physical robot efficiently reduces the  ...  INTRODUCTION In robotic system design and development, a large number of components must be considered simultaneously, such as the motor system, morphology, control policy, sensory apparatus, etc.  ... 
doi:10.1109/ssci.2016.7850169 dblp:conf/ssci/RuudSG16 fatcat:br73isxrrved7j65smhm4yytne

Revolve: A Versatile Simulator for Online Robot Evolution [chapter]

Elte Hupkes, Milan Jelisavcic, A. E. Eiben
2018 Lecture Notes in Computer Science  
In this paper, we present a system to simulate online evolution of constructible robots, where 1) the population members (robots) concurrently exist and evolve their morphologies and controllers, 2) all  ...  The necessary evolutionary mechanisms need not wait for these, they can be developed and investigated in simulations.  ...  Acknowledgement The choice of modular robots used for this research is based on the design of Josh Aurebach's RoboGen project -a flexible and scalable modular robot design.  ... 
doi:10.1007/978-3-319-77538-8_46 fatcat:oktnq66dgba2zbnjfjmnvj67wu

Introduction to the Evolution of Physical Systems Special Issue

John Rieffel, Jean-Baptiste Mouret, Nicolas Bredeche, Evert Haasdijk
2017 Artificial Life  
We deliberately chose this term to be broad enough to encompass both parallel Embodied Evolution [19] , in which evolution is distributed across a population of robots, as well as more classical Evolutionary  ...  Inspired by our mutual interests, we coined the term Evolution of Physical Systems (EPS) to describe evolutionary approaches that occur in real-world physical substrates rather than in simulation.  ...  A first idea is to start evolution in simulation then finish it on the real robot(s); this is, for instance, what Lipson and Pollack did to evolve the morphology and controllers of 3D-printed robots [  ... 
doi:10.1162/artl_e_00232 pmid:28513205 fatcat:gamjgzyx35gjdpeknfgzbwplcq

Real-world evolution adapts robot morphology and control to hardware limitations

Tønnes F. Nygaard, Charles P. Martin, Eivind Samuelsen, Jim Torresen, Kyrre Glette
2018 Proceedings of the Genetic and Evolutionary Computation Conference on - GECCO '18  
The large gap between performance in simulation and the real world makes it challenging to transfer the resulting robots to the real world.  ...  Most of the research in this ield, however, uses simpliied representations of the robotic system in software simulations.  ...  Figure 5 shows a large diversity in inal populations for the two groups of runs.  ... 
doi:10.1145/3205455.3205567 dblp:conf/gecco/NygaardMSTG18 fatcat:mdhm5faxn5efthjijf5icthnza

Using robots to investigate the evolution of adaptive behavior

Josh Bongard
2015 Current Opinion in Behavioral Sciences  
In the field of evolutionary robotics, investigators evolve populations of autonomous machines to exhibit some desired behavior.  ...  To illustrate this approach, I describe recent work on the evolution of modularity, morphological computation, and prospection. Address  ...  Because modularity is known to increase evolvability [17] , we observed that such robots outcompeted non-modular robots in the population, leading to the evolution of neural modularity in robots ( Figure  ... 
doi:10.1016/j.cobeha.2015.11.008 fatcat:5e7efvongzblxoyxp3k7r4guhy

Real-World Evolution Adapts Robot Morphology and Control to Hardware Limitations [article]

Tønnes F. Nygaard, Charles P. Martin, Eivind Samuelsen, Jim Torresen, Kyrre Glette
2018 arXiv   pre-print
The large gap between performance in simulation and the real world makes it challenging to transfer the resulting robots to the real world.  ...  Most of the research in this field, however, uses simplified representations of the robotic system in software simulations.  ...  Figure 5 shows a large diversity in final populations for the two groups of runs.  ... 
arXiv:1805.03388v1 fatcat:kwpmjidsvzbvjb7ef2ceretbqu

Overcoming Initial Convergence in Multi-objective Evolution of Robot Control and Morphology Using a Two-Phase Approach [chapter]

Tønnes F. Nygaard, Eivind Samuelsen, Kyrre Glette
2017 Lecture Notes in Computer Science  
Co-evolution of robot morphologies and control systems is a new and interesting approach for robotic design.  ...  Further, mutations in the robot morphologies tend to cause large perturbations in the search, effectively changing the environment, from the controller's perspective.  ...  However, there have also been some examples of evolution in simulation leading to instantiation of working physical robots [3, 6, 7] , and recently even simulationless evolution of robotic morphologies  ... 
doi:10.1007/978-3-319-55849-3_53 fatcat:zfcvu5hslfaynkoaqq2j7xftya

Evolutionary robotics in two decades: A review

2015 Sadhana (Bangalore)  
Principles of biological evolution are applied in the form of evolutionary techniques for solving the complicated problems in the areas of robotic design and control.  ...  The diversity and the intensity of this growing field is presented in this paper through the contributions made by several researchers in the categories of robot controller design, robot body design, co-evolution  ...  A chart is presented in figure 6 to sum-up the research presented in last two decades in this direction of co-evolution of robot morphology and control.  ... 
doi:10.1007/s12046-015-0357-7 fatcat:s7kkwrnsxzc75pinbjueukbpkq

Gait-learning with morphologically evolving robots generated by L-system [article]

Jie Luo, Daan Zeeuwe, Agoston E. Eiben
2021 arXiv   pre-print
We compare these two methods by running experiments in a simulator called Revolve and use efficiency, efficacy, and the morphology intelligence of the robots for the comparison.  ...  The experiments show that the evolution plus learning method does not only lead to a higher fitness level, but also to more morphologically evolving robots.  ...  Evolution + learning robots tend to be bigger along the evolution and wider in width compared to robots evolved in Evolution only.  ... 
arXiv:2107.08249v2 fatcat:4kwondw4p5czledcdgzsnrtgzi

Evolution of Adaptive Behaviour in Robots by Means of Darwinian Selection

Dario Floreano, Laurent Keller
2010 PLoS Biology  
Joint Evolution of Brains and Body Morphologies Experimental evolution has also been used to coevolve artificial brains and morphologies of simulated robots.  ...  Predator-Prey Coevolution Experimental evolution with robots has also been used to study the coevolutionary processes between a population of predator robots and a population of prey robots [28, 29] .  ... 
doi:10.1371/journal.pbio.1000292 pmid:20126252 pmcid:PMC2811146 fatcat:pujjqv4o4vf2fmtxhetsklyxpq

Cooperative co-evolution of configuration and control for modular robots

Chourouk Guettas, Foudil Cherif, Thomas Breton, Yves Duthen
2014 2014 International Conference on Multimedia Computing and Systems (ICMCS)  
The modular robots are evaluated in a simulation environment implemented with NVidia physics engine; PhysX.  ...  The general approach in modular robots is to hand design the morphology, and then optimizes the controller of the structure for a given task.  ...  ACKNOWLEDGMENT We specially thanks Yves Duthen, for facilitating our research by providing us with his developed modular robotics simulator and all the necessary documentations.  ... 
doi:10.1109/icmcs.2014.6911138 fatcat:55ft4x4w45aixadq33cvawngbi

Towards the targeted environment-specific evolution of robot components

Jack Collins, Wade Geles, David Howard, Frederic Maire
2018 Proceedings of the Genetic and Evolutionary Computation Conference on - GECCO '18  
simulator.  ...  The leg is represented in the simulator as 3D-printable file, and as such can be readily instantiated in reality.  ...  ER experiments require large numbers of iterations of a population of solutions; simulators used in ER therefore tend to emphasise speed over environmental realism and fidelity, being largely devoid of  ... 
doi:10.1145/3205455.3205541 dblp:conf/gecco/CollinsGHM18 fatcat:tfyp5l5dwrbc7g4vcn43ayisz4

A survey on evolutionary-aided design in robotics

Shanker G. Radhakrishna Prabhu, Richard C. Seals, Peter J. Kyberd, Jodie C. Wetherall
2018 Robotica (Cambridge. Print)  
This paper covers the efforts to determine body morphology of robots through evolution and body morphology with the controller of robots or similar creatures through co-evolution.  ...  EAs have found applications in many areas of robotics.  ...  This paper reports the evolution of the morphology alone along with the co-evolution of morphology and controller in the traditional areas of robotics.  ... 
doi:10.1017/s0263574718000747 fatcat:ryqytqaprbcxxizs7hz3qmhl7u

EvoSphere: The World of Robot Evolution [chapter]

A. E. Eiben
2015 Lecture Notes in Computer Science  
On the long term, EvoSphere technology can pave the way for robot populations that evolve 'in the wild' and can adapt to unforeseen and changing circumstances.  ...  In this paper I describe EvoSphere, a tangible realization of the general Evolution of Things concept.  ...  a population of robots that 'live' and reproduce in a simulated world [27] .  ... 
doi:10.1007/978-3-319-26841-5_1 fatcat:afancuzx5jcvfcqatoakovpm5a

The Utility of Evolving Simulated Robot Morphology Increases with Task Complexity for Object Manipulation

Josh Bongard
2010 Artificial Life  
The utility of evolving simulated robot morphology increases with task complexity for object manipulation. Artificial Life, uncorrected proof.  ...  An increasingly common approach therefore is to evolve an agent's morphology along with its control in the hope that evolution will find a good coupled system.  ...  Bongard The Utility of Evolving Simulated Robot Morphology Figure 3b reports the mean number of objects manipulated by robots in the population during that generation.  ... 
doi:10.1162/artl.2010.bongard.024 pmid:20059328 fatcat:odu4m7enjbbt5chfm7esimvuaq
« Previous Showing results 1 — 15 out of 6,398 results