Spatial, temporal, and modulatory factors affecting GasNet evolvability in a visually guided robotics task

Husbands, Phil; Philippides, Andrew; Vargas, Patrícia A.; Buckley, Christopher L.; Fine, Peter; Paolo, Ezequiel A. Di; O’Shea, Michael

doi:10.1002/cplx.20336

Cited by 20 publications

(22 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Both these extended forms proved to be significantly more evolvable again than the basic GasNet. Other varieties include nonspatial GasNets where the diffusion process is replaced by explicit gas connections with complex dynamics (Vargas et al 2009) and version with other forms of modulation and diffusion (Husbands et al 2010). In order to gain insight into the enhanced evolvability of GasNets, detailed comparative studies of these variants with each other, and with other forms of ANN, were performed using the robot task illustrated in Figure 3 ( Philippides et al 2005b.…”

Section: Volume Signalling: Gasnetsmentioning

confidence: 99%

See 1 more Smart Citation

Evolutionary Robotics and Neuroscience

Husbands

Moioli

Shim

et al. 2014

The Horizons of Evolutionary Robotics

View full text Add to dashboard Cite

Section: Volume Signalling: Gasnetsmentioning

confidence: 99%

“…Four of the nodes are assigned as motor nodes (forward and backward nodes for the left and right motor, with motor speeds proportional to the output of the relevant forward node minus the output of the relevant backward node). See Husbands et al (1998Husbands et al ( , 2010 for full details.…”

Section: Flatenmentioning

confidence: 99%

Evolutionary Robotics and Neuroscience

Husbands

Moioli

Shim

et al. 2014

The Horizons of Evolutionary Robotics

View full text Add to dashboard Cite

“…The next two articles consider the effects of spatial embedding on artificial and biological neural networks. Husbands et al [23] explore the influence of spatial and temporal constraints on a class of artificial neural networks known as GasNets, that incorporate two interacting signaling mechanisms: one based on electrical connections, the other on a diffusible gaseous neurotransmitter. Their findings indicate that spatial embedding facilitates exploitation of loose coupling between these signaling mechanisms in a way that enhances evolvability.…”

Section: Contents Of This Special Issuementioning

confidence: 99%

Spatial embedding as an enabling constraint: Introduction to a special issue of complexity on the topic of “Spatial Organization”

Bullock

Geard

2010

Complexity

View full text Add to dashboard Cite

We introduce and discuss the role of spatial embedding as an enabling constraint on complex system structure and function.Key words: complexity, space, spatial embedding, networks, Space as an Enabling ConstraintThe vast majority of real-world complex systems are extended in space. Some parts are close together, others are far apart. Moreover, the tendency for particular system parts to interact is partly determined by the distances that separate them. Consequently, there is considerable evidence that this spatial embedding has a significant impact on both the structural and functional organisation of complex systems. Spatial constraints are implicated in the way that populations of termites construct their impressive mounds [15,16], the way in which species of plants compete for light [8], and even the way in which genes are packed onto linear strands of DNA [18]. Moreover, we are beginning to understand that this influence is not neutral, but can sometimes be positive, with spatially embedded systems enjoying an improved ability to effectively organise in order to support useful functionality.Key examples are offered by [5] in their work modelling the tendency of molecules to organise into mutually self-reinforcing "hypercycles" [11], and by [9] in the context of trying to understand how biological populations of simple creatures might evolve to exhibit co-operative tendencies (see [6], this volume, for more details of both models). In each case, the system being Preprint submitted to Complexity April 22, 2010 modelled was unable to exhibit organised complex behaviour when its constituent parts were well-mixed. However, when each system was extended over a two-dimensional space such that its parts could only interact with their close spatial neighbours, the systems were able to spontaneously achieve and maintain functional organisation in the face of exploitative disturbance. Spatial embedding enabled each population (of modelled molecular species or simple simulated creatures) to structure itself in such a way as to maintain co-operative functional organisation through exploiting the "useful" asymmetries that space introduces into the ecology of interactions [7]. These results suggest an intriguing possibility: might understanding the role of spatial embedding help us to better deal with problems concerning a wide range of spatially embedded complex systems? For instance, how to organise the built environment such that community coherence is consolidated rather than fragmented [4]; how to arrange social care processes such that they are vital and integrated [17]; how to engineer accountability in virtual communities [12]; how to achieve bioremediation using bacterial biofilm communities [19]; and how to improve the resilience of the infrastructural systems upon which we rely [10], etc. Could the influence of spatial embedding in these contexts, when properly understood, allow and encourage individuals to organise in a way that makes good solutions easier to achieve?It is in this sense that space sh...

show abstract

“…This, combined with insights from evolutionary robotics (Nolfi et al, 2016;Vargas et al, 2014) where it was discovered that dynamically complex neural networks are highly evolvable (Husbands et al,2 Evolutionary Computation Volume x, Number x 2010; Beer and Williams, 2015), suggested that analog EHW might be very well suited to evolving compact controllers operating with small numbers of components, even for visually guided behaviours which traditionally employed high levels of processing (Bekey, 2005). An interesting property of the dynamically complex evolved networks mentioned above is their ability to cope with noisy, poor quality sensory data, even when the networks have very few nodes (Husbands et al, 2010). This suggests that analog EHW might also be a useful approach for low cost realtime hardware applications requiring cheap sensors and simple circuits.…”

Section: Introductionmentioning

confidence: 99%

Evolved Transistor Array Robot Controllers

Garvie

Flascher²,

Philippides

et al. 2020

Evolutionary Computation

Self Cite

View full text Add to dashboard Cite

For the first time a field programmable transistor array (FPTA) was used to evolve robot control circuits directly in analog hardware. Controllers were successfully incrementally evolved for a physical robot engaged in a series of visually guided behaviours, including finding a target in a complex environment where the goal was hidden from most locations. Circuits for recognising spoken commands were also evolved and these were used in conjunction with the controllers to enable voice control of the robot, triggering behavioural switching. Poor quality visual sensors were deliberately used to test the ability of evolved analog circuits to deal with noisy uncertain data in realtime. Visual features were coevolved with the controllers to automatically achieve dimensionality reduction and feature extraction and selection in an integrated way. An efficient new method was developed for simulating the robot in its visual environment. This allowed controllers to be evaluated in a simulation connected to the FPTA. The controllers then transferred seamlessly to the real world. The circuit replication issue was also addressed in experiments where circuits were evolved to be able to function correctly in multiple areas of the FPTA. A methodology was developed to analyse the evolved circuits which provided insights into their operation. Comparative experiments demonstrated the superior evolvability of the transistor array medium.

show abstract

Spatial, temporal, and modulatory factors affecting GasNet evolvability in a visually guided robotics task

Abstract: Spatial, temporal, and modulatory factors affecting the evolvability of GasNets

Cited by 20 publications

References 21 publications

Evolutionary Robotics and Neuroscience

Evolutionary Robotics and Neuroscience

Spatial embedding as an enabling constraint: Introduction to a special issue of complexity on the topic of “Spatial Organization”

Evolved Transistor Array Robot Controllers

Contact Info

Product

Resources

About