ARQUITECTURAS AUTOENSAMBLANTES
SELF-ASSEMBLY ARCHITECTURES
Abstract
Recent advancements in robotics have sparked an exploration of new ways to think aboutarchitecture. This project proposes the idea that it is possible to develop form-findingautonomous robotic systems with fractal geometries at its core. Self-assembly is a process whereby a robot or other object can construct itself by assembling smaller parts into a larger whole. This project aims to demonstrate that it is possible to develop physical form-finding autonomous systems, based on self-assembling programmable modules that interact with the environment to generate architectures that are adapted to specific conditions.
Keywords
Self-assembly, programmable matter, physical computing, modular design, form-finding, generative design, fractal growth.
In recent years, there has been an increasing interest in using robotics to rethink architecture. Traditionally, robotic systems have been used for tasks such as knitting, carving, stacking up modules or 3D printing structures. Recent advancements in robotics have sparked an exploration of new ways to think about architecture. This project proposes the idea that it is possible to develop form-finding autonomous robotic systems with fractal geometries at its core. Self-assembly is a process whereby a robot or other object can construct itself by assembling smaller parts into a larger whole. This process often takes place without the need for any external intervention, making it an attractive option for building and designing autonomous systems(Ekblaw and Paradiso 2018). This project aims to demonstrate that it is possible to develop physical form-finding autonomous systems, based on programmable modules that interact with the environment to generate architectures that are adapted to specific conditions.
The utilization of programmable modules with integrated magnets enabled the creation of a physical computing system capable of exhibiting a range of behaviors contingent upon the arrangement of the magnets. The system was implemented using two distinct programming approaches: the first utilizing a deterministic methodology (deterministic self-assembly), and the second utilizing a non-deterministic methodology (non-deterministic self-assembly). While the outcome of the first program, as anticipated, resulted in a predesigned pattern, the outcome of the second program was noteworthy as the resulting architectures exhibited a level of adaptation to the environmental conditions present within the experimental habitats.
The module
The initial conceptualization of the modules utilized a hexagonal tessellation strategy to create a system that could exhibit emergent behavior through local interactions that could be characterized as a pattern of incremental change or growth. Among the various possible configurations that could be constructed using six identical modules, the snowflake formation was deemed particularly noteworthy due to its symmetrical properties.
Figure 1 – Modules, patterns, and the snowflake pattern. (Own creation)
Deterministic Self-assembly
The initial study aimed to investigate the potential of using deterministic self-assembly in tangible and physical models. To do this, a specific structure resembling a snowflake was designed and constructed using six modular units, each of which was equipped with internal magnets arranged in a ring configuration (as shown in figures 2 and 3). Through the application of external forces or agitation, the system was able to be dismantled and subsequently reconfigured into a specific pattern. This demonstrated the feasibility of the system to reassemble into a predefined structure.
Figure 2 – Deterministic Self-assembly. (Own creation)
Figure 3 – QR Code for Deterministic Self-assembly video. (Own creation)
QR NOT INCLUDED YET SO THAT THE AUTHOR CANNOT BE IDENTIFIED
Evolution of the module
After the initial prototype of the modules was tested, a series of iterative design improvements were implemented to enhance the connectivity of the overall system. Various configurations were evaluated and systematically modified to ensure optimal compatibility and assembly with the other components of the system. Through a process of experimentation and refinement, the final geometry of the module was optimized to achieve maximum connectivity, as determined through objective measurements and performance evaluations.
Figure 4 – Evolution of the module. (Own creation)
The design process resulted in two different modules that were developed for the experiments. The basic module, which is the smallest component of the system and the snowflake module, which is based on the assemblage of six basic modules into the snowflake pattern. While the basic module has six slots, the snowflake has twelve slots that can be programmed with different magnets configurations and represents a second order of magnitude for the system.
| Basic modules(First order of magnitude) | Basic modules (Arranged in the snowflake pattern) | Snowflake module(Second order of magnitude) |
Figure 5 – Modules. (Own creation)
Programming
The use of magnets in the design of the modules allows for the programming of individual slots within each module with three distinct configurations: North, South, or Neutral (empty). This flexibility in module behavior is achieved by adjusting the magnetic orientation of the slots (as depicted in figures 6 and 11). This level of customization allows for the manipulation of connectivity levels within the self-assembly system. For example, the magnetic interactions between modules can be adjusted from high to low connectivity depending on the programmed configurations. This feature is of particular significance since it allows for greater control over the behavior of the system and the resulting structures.
Furthermore, this approach also allows for the creation of more complex structures, as the magnetic interactions between modules can be programmed to be specific, allowing for a higher degree of precision in the self-assembly process. Additionally, this method of controlling the interactions between modules may have potential applications in other fields as well, such as robotics or materials science. Overall, this approach adds a new dimension of control to self-assembly systems and expands the possibilities of what can be achieved with this technology.
| Universal Connector(High connectivity – low fragility) | V Connector | Ring Connector (Low connectivity – high fragility) |
Figure 6 –Basic Modules Programming. (Own creation)
In the top left corner of the following table, it is shown that two universal connectors can form nine different strong bonds (sb). Conversely, in the opposite corner, it is shown that two ring connectors can only form two weak bonds (wb). By utilizing magnets, the fragility and connectivity of the system can be easily programmed.
Table 1 – Matrix of possible bonds between modules
Non-deterministic self-assembly I
A 20-gallon fish tank was acquired to serve as an appropriate experimental habitat. Observations were conducted to assess the system’s capacity for growth and self-regulation, including the formation and dissolution of bonds among components, and the emergence of new patterns of organization. The system’s evolution towards increasingly stable configurations was examined, with a focus on identifying and analyzing patterns that reflect both stability and vulnerability. The goal of the experiment was to gain insight into the system’s tendency towards equilibrium within its environment.
Figure 7 – Experimental habitat before it is filled with water. (Own creation)
A structure designed to hold layers of air within the fish tank was implemented to achieve a comprehensive three-dimensional analysis of the environment, and turbulence to fabricate the experimental habitat was generated through the utilization of two hydro-jets. As depicted in figures 8 and 9, the proposed system utilizes a modular self-assembling construction methodology that adheres to the principles of physical dynamics, exhibits adaptability to its surrounding context, and continually reconfigures itself until a state of equilibrium is attained.
Figure 8 – Non-deterministic self-assembly I. (Own creation)
Figure 9 – QR Code for Non-deterministic self-assembly video (Own creation)
QR NOT INCLUDED YET SO THAT THE AUTHOR CANNOT BE IDENTIFIED
Attaining equilibrium refers to the state in which a system is in balance and no longer undergoing change. In the context of generating a comprehensive three-dimensional model, this equilibrium refers to the final configuration of a set of modules in a specific environment. Once this equilibrium state is reached, a scanning process can be applied to the resulting spatial configuration to generate a comprehensive three-dimensional model. This model captures the unique characteristics of a structure that is adapted to its environment, providing information about the form that a specific architecture designed for such environment could have.
This resulting architecture is tailored to the specific conditions of that environment, meaning that it is optimized to perform well in the given setting. This approach could be particularly useful in the fields of architectural design and civil engineering, as it allows for the creation of structures that are well-suited to the specific conditions of the site on which they are built. Additionally, this approach can be applied to other fields such as industrial design and robotics.
Figure 10 – 3D scanned model of resulting configuration. (Own creation)
Finally, it is important to mention that in the various experimental studies conducted, the connectivity and fragility of the system were manipulated to identify optimal configurations of modules that would yield comprehensive and cohesive architectures.
Non-deterministic self-assembly II
The second experiment investigating non-deterministic self-assembly was conducted in a lake, rather than a laboratory setting. This experiment utilized snowflake modules, which are larger and better able to withstand harsher environmental conditions. The outcome of the experiment was consistent with the hypothesis, with the system evolving towards stable configurations. However, it should be noted that due to the vastness of the lake environment, the system did not reach a state of equilibrium.
| Universal Connector (Snowflake) | Y Connector |
Figure 11 – Snowflake Modules Programming. (Own creation)
This experiment highlights the importance of considering the scale and complexity of the environment when studying self-assembly systems. It also demonstrates the potential for self-assembly to occur in natural environments, and not just in laboratory settings.
Figure 12 – Non-deterministic self-assembly II. (Own creation)
The use of larger modules in this experiment also indicates the potential for self-assembly at different scales and orders of magnitude, which is pivotal for depicting the possibility assemblages that follow fractal patterns. Overall, this experiment adds to the growing body of research on non-deterministic self-assembly and expands our understanding of the behavior of these systems in different environments.
The coming future of fractal self-assembly robotics
Self-assembly is a process in which individual components autonomously organize themselves into a specific structure or pattern without the need for external guidance or control. This concept has been applied to various fields such as materials science, nanotechnology, and robotics. The application of self-assembly to robotic systems can lead to the development of swarm robots that can assemble from a set of pre-programmed modules. However, the control mechanisms of the assembly process become a challenge as the system grows in complexity.
Fractal self-assembly has been successfully implemented at the nanoscale to create structures ranging from simple two-dimensional patterns to complex three-dimensional architectures(Lomander, Hwang, and Zhang 2005; Sarkar et al. 2014; Tikhomirov, Petersen, and Qian 2017). This project seeks to arouse interest in the study of fractal self-assembly to the field of architecture robotics.
This project suggests a shift in the way we currently understand self-assembly systems, specifically regarding the possibility of different orders of magnitude existing within the same system. Instead of traditional methods where cubes pile against each other, the proposed approach advocates for a paradigm that is more in line with the way natural systems, such as plants and organisms, grow and organize themselves – through fractal growth.
Figure 12 – Fractal Self-assembly
Fractal growth refers to the process of self-similar patterns being repeated at different scales, resulting in a hierarchical structure. In the context of self-assembly systems, this would mean considering the interactions and organization of components at multiple levels or orders of magnitude. This approach could potentially lead to a more efficient and effective way of designing and implementing self-assembly systems, as it considers the complex and dynamic nature of natural systems.
Recent advancements have introduced magnetically reprogrammable materials which can be pivotal to make these ideas possible in the future, enabling fractal self-assembly into specific shapes and chosen configurations (Nisser et al. 2022) These advances could allow us to better mimic nature, to produce better machines that follow the mathematical logic of growth in nature.
Finally, by exploring the use of fractal growth in self-assembly systems, this research aims to contribute to the broader field of complex systems, by providing a new perspective on how to understand and analyze the organization and behavior of systems that are composed of multiple components. Overall, this research proposes to move towards a new paradigm in thinking about self-assembly systems that is more in line with the way nature grows and organizes itself, by embracing the concept of fractal growth.
References
Ekblaw, Ariel, and Joseph Paradiso. 2018. “TESSERAE: Self-Assembling Shell Structures for Space Exploration.” In Proceedings of IASS Annual Symposia, 2018:1–8. International Association for Shell and Spatial Structures (IASS).
Lomander, Andrea, Wonmuk Hwang, and Shuguang Zhang. 2005. “Hierarchical Self-Assembly of a Coiled-Coil Peptide into Fractal Structure.” Nano Letters 5 (7): 1255–60.
Nisser, Martin, Yashaswini Makaram, Faraz Faruqi, Ryo Suzuki, and Stefanie Mueller. 2022. “Selective Self-Assembly Using Re-Programmable Magnetic Pixels.” ArXiv Preprint ArXiv:2208.03799.
Sarkar, Rajarshi, Kai Guo, Charles N Moorefield, Mary Jane Saunders, Chrys Wesdemiotis, and George R Newkome. 2014. “One‐Step Multicomponent Self‐Assembly of a First‐Generation Sierpiński Triangle: From Fractal Design to Chemical Reality.” Angewandte ChemieInternational Edition 53 (45): 12182–85.
Tikhomirov, Grigory, Philip Petersen, and Lulu Qian. 2017. “Fractal Assembly of Micrometre-Scale DNA Origami Arrays with Arbitrary Patterns.” Nature 552 (7683): 67–71.
Alejandro Puentes Amézquita | Architectures 