RESEARCH
COAXIAL RESEARCH platform proposes the Theory of Evolution for Architectural Intelligent Adaptive Systems. The computational architectural language generates intelligent architectural structures that achieve balanced environmental conditions for individuals and communities based on their space experience, sensing and behaviour. The intelligence is encoded in a script of an algorithms, neural networks models that are capable of rewriting their existing code protocols, and therefore actively forward demands on architecture for effective and dynamic adaptability. The major challenge faced by Architectural Intelligence is to develop expert agents that can learn to perform many different tasks and learn with the little domain-specific knowledge in real-time.
The research field investigates the applicability of machine learning algorithms in the context of spatial dynamic forces, as a tool for architectural representation of sound and dynamic spatial structures. The thesis deals with the application of more complex machine learning models for Architectural Intelligence.
Architectural Intelligence is a set of evolutionary mechanisms that has a capability to adapt the architectural organism to the new environmental situation or behavioural patterns of it’s symbionts, in a sort-term or long-term interactions. The Architectural Intelligence is both adapting, assimilating and accommodating the environmental dynamic and behavioural conventions. The architectural intelligence is taught by its architect. Regarding Piaget intelligence can be seen as a form of adaptation in which knowledge is constructed by each individual through two complementary processes of assimilation and accommodation (Jean Piaget, 1963). The architectural adaptation is an evolutionary process as a result of which the architectural body better adapts to a dynamically changing environment. If an organism cannot move or change enough to maintain its long-term viability, it will vanish.
Architectural Intelligence model of neuro-evolution and meta-learning models provides variability and flexibility in dynamic environments. Meta-learning approach provide sustainable possibility of implementing already tested and trained models from other domains and areas of machine learning to the field of architecture. The above research is fundamental to an architecture of the future that will be well adapted, to the changes defined as differentiations of the environment, in particular resilient architecture that accommodates mass migrations and environmental climate crisis situations or global pandemics.
Neuro-Architecture
Neuro-Architecture represents a groundbreaking convergence of architecture, neuroscience, and neuro-evolutionary theory. This multidisciplinary approach aims to develop a new architectural language that not only communicates with our technologically advanced and information-dense world, but also encapsulates the complex behavioural and environmental qualities inherent in architectural neuro-material. This innovative language of architecture serves as a dynamic medium, translating intricate patterns of human behaviour and environmental interactions into the physical form of architectural spaces. Through this transformative process, Neuro-Architecture seeks to redefine our built environments, creating spaces that are not just structures, but living entities that interact, adapt, and evolve with us.
Evolutionary-Achitecture
The early stage of research experimentation with the simple neural networks and basic machine learning technics, outlined a need of more general framework for optimization and adaptation of architectural space. The basic generative
methods as a results of the data processing from the digital environment could not reach the complexity of the proposed framework of Quantum model for Architectural Intelligence. The genetic and evolutionary algorithms had to be
applied to the model.
Evolutionary-Architecture proposes a new strategy for evolving architectural structures based on the idea of adaptation to a dynamically changing environment, with the use of advanced machine learning and AI methods. The thesis explores
a living dynamic system as a complex set of natural and cultural sub-processes, in which each of the interacting entities and systems creates complex aggregates.
Meta-Architecture
It is a way of designing in the design, where theory and practice, criticism and creation, thinking and doing coincide. Meta-architecture is also a way of exploring the dynamic environment, which is a model in virtual world that
transcends the boundaries of physical reality. The research by Kotnour is an example of meta-architecture, as it uses machine learning and AI methods to create and evolve architectural structures that adapt to a dynamic environment,
such as soundscapes. This field of research also uses cognitive neuroscience and human-computer interaction to transfer knowledge from other disciplines into architectural creation. The research aims to develop a new form of Architectural
Intelligence, which is the ability to design, learn, and evolve in complex and uncertain situations and environmental scenarios.