Exploring Natural Strategies for Bio-Inspired Fault Adaptive Systems Design

Arroyo, Marvin; Huisman, Nicholas; Jensen, David

doi:10.31224/osf.io/w8vng

Cited by 2 publications

(3 citation statements)

References 10 publications

(10 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Much research has been done so far to study ways of designing complex systems. For instance, Arroyo et al (2018) came up with a bio-inspired, binary-tree-based method for complex system design with fault adaptiveness. The binary-tree branches indicate how organisms achieve adaptive behaviors.…”

Section: Related Workmentioning

confidence: 99%

“…The binary-tree branches indicate how organisms achieve adaptive behaviors. Their results show that a strategy-based method can aid designers with innovative analogies, which is useful in providing needed details in design applications (Arroyo et al , 2018). Königseder and Shea (2016) analyzed various strategies of rule applications in two design synthesis case studies: bicycle frame synthesis and gearbox synthesis, and they found that the effectiveness of each strategy depends on the task scenario.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating the learning and performance characteristics of self-organizing systems with different task features

Jin

2021

AIEDAM

View full text Add to dashboard Cite

Self-organizing systems (SOS) are developed to perform complex tasks in unforeseen situations with adaptability. Predefining rules for self-organizing agents can be challenging, especially in tasks with high complexity and changing environments. Our previous work has introduced a multiagent reinforcement learning (RL) model as a design approach to solving the rule generation problem of SOS. A deep multiagent RL algorithm was devised to train agents to acquire the task and self-organizing knowledge. However, the simulation was based on one specific task environment. Sensitivity of SOS to reward functions and systematic evaluation of SOS designed with multiagent RL remain an issue. In this paper, we introduced a rotation reward function to regulate agent behaviors during training and tested different weights of such reward on SOS performance in two case studies: box-pushing and T-shape assembly. Additionally, we proposed three metrics to evaluate the SOS: learning stability, quality of learned knowledge, and scalability. Results show that depending on the type of tasks; designers may choose appropriate weights of rotation reward to obtain the full potential of agents’ learning capability. Good learning stability and quality of knowledge can be achieved with an optimal range of team sizes. Scaling up to larger team sizes has better performance than scaling downwards.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Evaluating the learning and performance characteristics of self-organizing systems with different task features

Jin

2021

AIEDAM

View full text Add to dashboard Cite

show abstract

“…Adaptive System Design tool, etc., to also incorporate possible evolutionary constraints, in order to arrive at not only the best possible BID solution, but also at identifying potentials for further optimization [25,31,[88][89][90][91].…”

Section: Evomimetic Design Thinkingmentioning

confidence: 99%

Evomimetics: the biomimetic design thinking 2.0

Adriaens

2019

Bioinspiration, Biomimetics, and Bioreplication IX

View full text Add to dashboard Cite

The consensus is that nature is a tremendous source of ideas for innovative designs that can meet various specific functional needs, relevant to society. Designs rely on structural, constructional, process-based and behavioral traits that all result from a natural trial-and-error cycle: evolution. Being one of the pillars of biomimicry, through billion years of evolution, nature has experimented and found what works and lasts, and what does not. Evidently, this has attracted scientists, especially engineers, trying to understand working natural designs, and translate them into applicable, working synthetic designs. The 'Biomimetic Design Method' forms the underlying conceptual framework to analytically decode biologically functions and designs. However, even though the evolutionary process is considered key to all this, it is generally overlooked in this conceptual thinking. The general assumption is that particular functions in organisms result from a natural selection process that optimized the underlying design for a particular function, thereby overlooking that an organism actually represents the possibly best compromise between all its functions needed to survive, to reproduce and to produce fit offspring. Many evolutionary processes thus yield suboptimal design components that, when put together, provide an optimized organismal design that manages to perform as good as needed, within a given environment. Such evolutionary limitations thus create possible pitfalls for bio-inspired design thinking. But, when considering them as a structural part of the design thinking process ('evomimetics'), they actually create opportunities for an improved translation of biology into optimally functioning designs. Using specific examples from evolutionary biology, these processes are explained, and recommendations are formulated.

show abstract

Exploring Natural Strategies for Bio-Inspired Fault Adaptive Systems Design

Cited by 2 publications

References 10 publications

Evaluating the learning and performance characteristics of self-organizing systems with different task features

Evaluating the learning and performance characteristics of self-organizing systems with different task features

Evomimetics: the biomimetic design thinking 2.0

Contact Info

Product

Resources

About