Evolution and Development of Modular Control Architectures for 1D Locomotion in Six-legged Animats

Kodjabachian, Jérôme; Meyer, Jean-Arcady

doi:10.1080/095400998116413

Cited by 30 publications

(25 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Results concerning the automatic production of 1-Dlocomotion controllers have been reported at length elsewhere [13]. In particular, it has been shown that some of the neural networks that have been obtained were capable of generating a tripod gait because they called upon four central pattern generators that were responsible for the rhythmic movements of the middle and back legs.…”

Section: A 2-d-locomotionmentioning

confidence: 94%

See 1 more Smart Citation

Evolution and development of neural controllers for locomotion, gradient-following, and obstacle-avoidance in artificial insects

Kodjabachian

Meyer

1998

IEEE Trans. Neural Netw.

123

View full text Add to dashboard Cite

This paper describes how the SGOCE paradigm has been used to evolve developmental programs capable of generating recurrent neural networks that control the behavior of simulated insects. This paradigm is characterized by an encoding scheme, by an evolutionary algorithm, by syntactic constraints, and by an incremental strategy that are described in turn. The additional use of an insect model equipped with six legs and two antennae made it possible to generate control modules that allowed it to successively add gradient-following and obstacle-avoidance capacities to walking behavior. The advantages of this evolutionary approach, together with directions for future work, are discussed.

show abstract

Section: A 2-d-locomotionmentioning

confidence: 94%

“…In [13], Manuscript we have implemented such a developmental process within the framework of an incremental evolutionary approach that made it possible to evolve one-dimensional (1-D) locomotion controllers in six-legged animats, i.e., in animats that walk along straight lines.…”

Section: Introductionmentioning

confidence: 99%

Evolution and development of neural controllers for locomotion, gradient-following, and obstacle-avoidance in artificial insects

Kodjabachian

Meyer

1998

IEEE Trans. Neural Netw.

123

View full text Add to dashboard Cite

show abstract

“…Their networks grow in a two-dimensional space using processes such as cell division and axon growth. Kodjabachian and Meyer also used connection growth mechanisms in their geometry-oriented version of CE called SGOCE [29]. Utilizing similar ideas of development, Miller evolved developmental programs that can construct the French flag (i.e.…”

Section: B Indirect Encodingsmentioning

confidence: 99%

Evolving Symmetry for Modular System Design

Valsalam

Miikkulainen

2011

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

Abstract-Symmetry is useful as a constraint in designing complex systems such as distributed controllers for multilegged robots. However, it is often difficult to determine which symmetries are appropriate. It is therefore desirable to design such systems automatically, e.g. by utilizing evolutionary algorithms that produce symmetry through developmental mechanisms. The success of these algorithms depends on how well they explore the space of valid symmetries. This paper presents an approach called Evolution of Network Symmetry and mOdularity (ENSO) that utilizes group theory to search the space of symmetries effectively. This approach was evaluated by evolving neural network controllers for a quadruped robot in physically realistic simulations. On flat ground, the resulting controllers are as fast as those having hand-designed symmetry, and significantly faster than those without symmetry. On inclined ground, where the appropriate symmetries are difficult to determine manually, ENSO produced significantly faster gaits that also generalize better than those of other approaches. On robots with a more complicated structure including knee joints, ENSO resulted in more regular gaits than the other approaches. These results suggest that ENSO is a promising approach for evolving complex systems with modularity and symmetry.

show abstract

“…We created three new developmental models initially inspired by Kodjabachian and Meyer's SGOCE paradigm [26][27][28][29][30] and NEAT [24,25,31]. By using them, we wanted to see how bilateral symmetry in neural networks could be generated and affect the behavior of a simulated robot.…”

Section: Our Approachmentioning

confidence: 99%

Evolution of bilateral symmetry in agents controlled by spiking neural networks

Oros¹,

Steuber²,

Davey³

et al. 2009

2009 IEEE Symposium on Artificial Life

View full text Add to dashboard Cite

Abstract-We present in this paper three novel developmental models allowing information to be encoded in space and time, using spiking neurons placed on a 2D substrate. In two of these models, we introduce neural development that can use bilateral symmetry. We show that these models can create neural controllers for agents evolved to perform chemotaxis. Neural bilateral symmetry can be evolved and be beneficial for an agent. This work is the first, as far as we know, to present developmental models where spiking neurons are generated in space and where bilateral symmetry can be evolved and proved to be beneficial in this context.

show abstract

Evolution and Development of Modular Control Architectures for 1D Locomotion in Six-legged Animats

Cited by 30 publications

References 25 publications

Evolution and development of neural controllers for locomotion, gradient-following, and obstacle-avoidance in artificial insects

Evolution and development of neural controllers for locomotion, gradient-following, and obstacle-avoidance in artificial insects

Evolving Symmetry for Modular System Design

Evolution of bilateral symmetry in agents controlled by spiking neural networks

Contact Info

Product

Resources

About