Search

While the utilization of robot arms has increased since the construction industry began to deploy robotic technologies for digital fabrication processes, a pipeline is missing for fabrication-aware design as the abstraction of complex, contradictory constraints for the designer is not evident. %the non-standard characteristics of building components still pose major challenges that require flexible and adaptable robotic fabrication strategies. Additional geometric complexity, material properties, etc. also contribute to the overall difficulties for fabricating the designated piece successfully without any collisions or structural failure. Through the development of two projects focusing on different aspects of robotic fabrication, this dissertation identifies various limitations related to the overall design-to-fabrication process and categorizes them into different types of constraints. It is observed that many of the constraints occurred within one fabrication task are usually intertwined and cannot be decoupled, which requires integrated computational strategies to resolve. By adopting available methods in the computer graphics field that address geometry and material, this dissertation presents a series of optimization-based strategies in the context of two specific research projects, targeting geometry processing and path planning for robotic fabrication. Its aim is to demonstrate the potential of using optimization methods to obtain achievable robotic fabrication solutions under sophisticated requirements. Focusing on geometry processing and path planning, respectively, this dissertation employs optimization approaches to assist with design aims, and develops a conceptual framework for solving fabrication-aware robotic fabrication tasks. The formulation of the optimization problems in this dissertation empowers the design processes to be fabrication-aware so as to be compatible with the selected fabrication technology. It provides a more mathematical and holistic perspective for looking at robotic fabrication technologies in the architectural domain.

While the global impact of plastic waste is increasingly concerning, the application of reused materials in the built environment remains little explored. This paper presents research into the reuse of plastic in architecture by means of computational design and robotic fabrication. Design possibilities using reclaimed plastic artefacts were explored by testing their structural stability and robotically modifying them in order to create a pavilion. While the design conceptualization started with the reclaimed material and the analysis of its potential, the digital workflow involved generative and performance- driven design, structural optimization and geometry generation for robotic fabrication.

This paper presents an integrated raw-wood fabrication workflow using an industrial robot arm and a laser scanner. The research is situated in a Swiss mountain forestry context where timber in its natural form could be applied locally without relying on large centralized timber industries. While local saw-mills relies on the standard processing tools to transform raw-wood into regular beams and boards, an automated robotic application could exploit timber in its natural form directly for construction. The research proposes a digital fabrication workflow that links industrial robot arm and laser scanning to adapt timber joinery tool-paths to irregular raw-wood such as straight, bent and forked tree trunks. The application employs ABB IRB6400 robot and Faro Focus S150 laser scanner. Several methods are developed for the point-cloud processing including flat-cut sectioning, mesh reconstruction and cylinder fitting. The methodology is tested in two stages: a) scanning a set of tree trunks b) continuous integration within robotic fabrication. The findings compare the scanning method in regards to calibration, pointcloud processing, time needed to communicate between the robot controller, the accuracy of the application. The results show that is possible to perform robotic cutting and scanning interchangeably and the speed of the scanning application is proportionally fast (3-4 min) considering the overall robotic cutting time (60-90 min) depending on the study case.

‘Imprimer la Lumière’ examines the making of a bioluminescent micro architecture. The project positions itself inside a sustainability agenda. By exploring the use of light-emitting bacteria as a material for architecture it asks what are the concepts, methods and technologies needed for designing with living materials. The project devises new means by which to design with the luminescent vibrio fischeri bacteria in a 3D printing manufacturing process based on extrusion principles. By combining the study of these living organisms and their appropriation through advanced robot-controlled 3D printing technologies, we establish a conceptual, material and technological framework for a bio-controlled bacteria growth and 3D extrusion process and a printable material based on agarose and gelatine. &nbsp

Urban Microclimate Canopy is a digitally fabricated fiber glass structure supporting climbing plants in order to explore new ways of integrating vegetation in densely built urban environments. A prototype was designed and manufactured in the context of an interdisciplinary studio with master&rsquo, s students following an approach of research by design. Varying the assembly of winding frames and fiber weaving syntax generates diverse geometric shape and structural performance. For two short-term exhibitions, ivy plants were temporarily installed in the structure. This first step was followed with a reflection of systematic integration of the growth processes of climbing plants and parametric design. An iterative solution is given, consisting of a feedback loop linking the design of the technical structure, the simulation of plant growth, and the simulation of the environmental effects of the hybrid structure. To achieve this a novel framework for simulating twining plant&rsquo, s growth on network-like structures is presented: external stimuli define a cone-shaped circumnutation space (searching space model) which results in a climbing path (climbing steps model). The framework is constructed to integrate improved individual functions (such as stimuli of circumnutation) for better simulation results. To acquire more knowledge about interactions between the plants and the fiber structure, the prototype was installed permanently and planted with three different climbing plants, representing different climbing mechanisms.

Developments in the automation of construction processes, observable in recent years, is focused on speeding up the construction of buildings and structures. Additive manufacturing using concrete mixes are among the most promising technologies in this respect. 3D concrete printing allows the building up of structure by extruding a mix layer by layer. However, the mix initially has low capacity to transfer loads, which can be particularly troublesome in cases of external components that need to be placed on top such as precast lintels or floor beams. This article describes the application of additive manufacturing technology in the fabrication of a building wall model, in which the door opening was finished with automatic lintel installation. The research adjusts the wall design and printing process, accounting for the rheological and mechanical properties of the fresh concrete, as well as design requirements of Eurocode. The article demonstrates that the process can be planned precisely and how the growth of stress in fresh concrete can be simulated, against the strength level developed. The conclusions drawn from this research will be of use in designing larger civil structures. Furthermore, the adverse effects of concrete shrinkage on structures is also presented, together with appropriate methods of control.

The large-scale urbanisation taking place in the developing world requires the construction industry to adopt alternative non-conventional renewable materials to reduce the unsustainable level of greenhouse gas emissions associated with the production of industrialised building materials. Bamboo is one of the most promising non-conventional building materials endemic to most developing countries, but there is still insufficient consistent information on the physical and mechanical properties of the numerous species suitable for construction. This study shows the potential of robotic fabrication to accelerate testing programmes on small clear samples of bamboo required to compare physical and mechanical properties across different species and differing plantation management practices. This fabrication method is applied on an experimental testing programme to determine the characteristic values of density, compressive strength, elastic modulus and shear strength ofPhyllostachys pubescens(moso),Guadua angustifoliaKunth (guadua) andGuadua angustifolia(oldhamii). The efficient development of comprehensive experimental datasets of clear samples of bamboo is fundamental to inform the development of future design guidelines for bamboo as a construction material.

vConcrete is a material favored by architects and builders alike due to its high structural strength and its ability to take almost any form. However, to shape concrete structures, heavy-duty formwork is usually necessary to support the fresh concrete while curing. To expand geometrical freedom, three-dimensional (3D) printed concrete formwork has emerged as a field of research. This article presents one possible application, a novel fabrication process that combines large-scale robotic fused deposition modeling 3D printing with simultaneous casting of a fast-hardening, set-on-demand concrete. This fabrication process, known as ‘‘Eggshell,’’ enables the production of nonstandard concrete structures in a material-efficient process. By casting a fast-hardening concrete in a continuous process, lateral pressure exerted by the fresh concrete is kept to a minimum. In this way, a 1.5-mm-thin thermoplastic shell can be used as a formwork, without any additional support. Geometries of different scales are tested in this article to evaluate the feasibility of the Eggshell fabrication process in an architectural context. An array of printing materials are also tested, and several different reinforcement concepts are analyzed. The findings are used to produce a full-scale architectural demonstrator project. This article shows that a wide range of concrete geometries can be produced in a material-efficient fabrication process, paving the way toward mass customization and structural optimization within concrete architecture., 3D Printing and Additive Manufacturing, 7 (2), ISSN:2329-7662, ISSN:2329-7670

The development of computational design technologies and prefabrication systems have enabled the construction of bespoke long-span spatial structures. However, the construction of such structures still relies on wasteful milling processes for the production of custom parts and labor-intensive processes for their manual assembly. Building upon prefabrication systems, several institutions investigated robotic processes for the automatic construction of bespoke spatial structures. However, the new challenges introduced by these complex processes have been only handled through inefficient and project-specifc fabrication strategies that lead to constrained designs. This thesis investigates computational design methods to tackle two of the most relevant challenges of robotically assembling spatial structures, which include the generation of collision-free robot paths and the handling of tolerance build-up. The two methods enable the computational rationalization of spatial structures, meaning that they allow verifying input designs on their buildability. Such verification is pursued through two complementary strategies. The first strategy is computational post-rationalization and allows verifying a design after it is defined. The second strategy is computational co-rationalization and allows re-adjusting a design while verifying its buildability. The ultimate goal of this thesis is to extend the range of spatial structures that can be robotically fabricated through efficient and less wasteful construction processes. An additional goal is to enable the computational rationalization of the structure ahead of the construction phase to explore a wider range of spatial structures. The investigation complements the investigation of other research projects, by integrating the methods researched by this thesis within the design workflow of these projects. This integration allows validating the methods through the computational rationalization of large-scale spatial structures and their realization in the Robotic Fabrication Laboratory at ETH Zurich.

We present a method for effectively planning the motion trajectory of robots in manufacturing tasks, the tool paths of which are usually complex and have a large number of discrete time constraints as waypoints. Kinematic redundancy also exists in these robotic systems. The jerk of motion is optimized in our trajectory planning method at the meanwhile of fabrication process to improve the quality of fabrication. Our method is based on a sampling strategy and consists of two major parts. After determining an initial path by graph search, a greedy algorithm is adopted to optimize a path by locally applying adaptive filers in the regions with large jerks. The filtered result is obtained by numerical optimization. In order to achieve efficient computation, an adaptive sampling method is developed for learning a collision-indication function that is represented as a support-vector machine. Applications in robot-Assisted 3-D printing are given in this article to demonstrate the functionality of our approach. Note to Practitioners-In robot-Assisted manufacturing applications, robotic arms are employed to realize the motion of workpieces (or machining tools) specified as a sequence of waypoints with the positions of tool tip and the tool orientations constrained. The required degree of freedom (DOF) is often less than the robotic hardware system (e.g., a robotic arm has six-DOF). Specifically, rotations of the workpiece around the axis of a tool can be arbitrary (see Fig. 1 for an example). By using this redundancy, i.e., there are many possible poses of a robotic arm to realize a given waypoint, the trajectory of robots can be optimized to consider the performance of motion in velocity, acceleration, and jerk in the joint space. In addition, when fabricating complex models, each tool path can have a large amount of waypoints. It is crucial for a motion planning algorithm to compute a smooth and collision-free trajectory of robot to improve the fabrication quality. The time taken by the planning algorithm should not significantly lengthen the total manufacturing time; ideally, it would remain hidden as computing motions for a layer can be done while the previous layer is printing. The method presented in this article provides an efficient framework to tackle this problem. The framework has been well tested on our robot-Assisted additive manufacturing system to demonstrate its effectiveness and can be generally applied to other robot-Assisted manufacturing systems.

This project explores the implementation of additive manufacturing for monolithic shells based on the deposition of different clay mixes through robotic spraying over a temporary fabric formwork. The framework sits at the intersection of ancient construction traditions, digital modelling techniques, and robotic control protocols. Raw mud can be found in ancient construction traditions, from mud huts to wattle and daub techniques, to the recent appearance of 3d printed extrusions in several academic research projects. Thin shell vaulting has relevant precedents throughout the 20th century, incorporating a variety of formwork methods for spray concrete ranging from pneumatics to fabrics to the recent fabric removable formwork and metal mesh reinforcement for robotic spray concrete (Veenendaal and Block, 2014). This highly sustainable project is actively implementing the principle of “resistance through form” (Eladio Dieste, 1996) at the design stage, by using geometries with curvatures that result in active surfaces that are optimized and tested in 3d modelling softwares (Rhino, Grasshopper and Karamba).&nbsp

The digital formulation of monolithic shell structures present some challenges related to the interface between computational design and digital fabrication techniques, such as the methods chosen for the suitable parametrization of the geometry (Bravo, M., 2013) [1], the optimization criteria of variables, the translation of the relevant code for digital fabrication, and especially the phasing related to the matter becoming material within the digital fabrication process. This paper exposes the process for the implementation of new digital and physical methods for monolithic shells design and construction using digital fabrication techniques combined with enjoyable manual craft. It explores the potential of carefully recalibrating the phasing between digital parametric modeling formulation and simulations with the physical translation aiming at renewing shells construction processes, in terms of the time it takes to build them, the optimization of materials and labor, and the feasibility of the dweller to be involved during different phases of construction. In addition, it will highlight significant changes that digital fabrication can engender in the use and the resulting aesthetics. (Huijben, Van Herwijnen, Nijsse, 2011) [2]. The paper will feature 2 case studies, detailing the steps required to take advantage of the circular data flow involved in digital modeling techniques, matter characteristics, structure distortion, and digital fabrication tools using robotics (Block, Veenendaal, 2015) [3]. It will explore examples implemented with an easily mounted temporary or lost formwork, and the methods of 3d scanning iteratively or in real time, so the form in progress can inform the system when at a later stage the fabrication tools will be depositing the material. The potential of integrating sensors and radars used in the car industry to create new protocols of construction will be explored, such as having a set rule between the angle of extrusion and the level of material’s viscosity. New interfaces between 3d modeling programs will be presented (such as rhino and robotic fabrication -Kuka prc for example-) and the resulting novel monolithic shell construction experiments will be revealed.&nbsp

This research paper presents a novel method for robotic spraying of glass-fibre reinforced concrete (GFRC) on a permeable reinforcement mesh. In this process, the mesh acts as a functional formwork during the concrete spraying process and as reinforcement once the concrete is cured, with the goal of producing slender reinforced concrete elements efficiently. The proof of concept presented in this paper takes inspiration from "Ferrocement" technique, developed in the 1940s by Pier Luigi Nervi (Greco, 1994) and shows how robotic spraying has the potential of producing such slender and bespoke reinforced concrete elements while also having the potential of reducing manual labour, waste and excess material. The system is coined with the name "Robotic AeroCrete" (or RAC) in reference to the use of an industrial robotic setup and the pneumatic projection of concrete., Architecture in the Age of the 4th Industrial Revolution – Proceedings of the 37th eCAADe and 23rd SIGraDi Conference

Diese Diplomarbeit untersucht die Bedeutung der digitalen Fabrikation und robotische Holzkonstruktionen anhand des historischen Hintergrunds der digitalen Fabrikation. Verschiedene Methoden der digitalen Fertigung werden erforscht, analysiert und aufgrund ihrer Vor- und Nachteile bewertet. Denn seit langer Zeit-CNC-Fr��sen oder -Schneiden war bereits in der Industrie im Einsatz und fand seinen Weg in die Architektur. Auch in der Automobilindustrie wurden Roboterarme eingesetzt, die jedoch darauf programmiert waren, ein Leben lang einen Job zu erledigen. Das rechnergest��tzte Design erm��glichte die Verwendung dieser flexiblen und pr��zisen Werkzeuge zur Herstellung von Freiformk��rpern. Roboterarme haben eine Vielzahl von Vorteilen, wie beispielsweise die Flexibilit��t in ihrer Bewegung, da sie nicht in einem kubischen Arbeitsraum arbeiten. Neben der Tatsache, dass sie nicht nur an einem Werkzeug befestigt sind, kann der Endeffektor jede Art von Arbeitswerkzeug sein. Der Roboterarm kann jede digitale Fertigungsmethode ausf��hren, angefangen von der additiven Methode bis zur subtraktiven Methode. Dies h��ngt vom Endeffektor ab, der am Roboter angebracht ist. Das rechnerische Design ist ein gro��er Meilenstein, da mit dem Roboterarm komplexe Formen und Geometrien hergestellt werden k��nnen. Das nat��rliche Bauprodukt Holz ist ein Material, das f��r die rechnergest��tzte Konstruktion nicht einfach zu handhaben ist, aber Forscher auf der ganzen Welt haben verschiedene neue Methoden zur Herstellung von Robotern erfunden, die die rechnergest��tzte Konstruktion von Holzprodukten erm��glichen. In dieser Forschungsarbeit werden das rechnergest��tzte Design von Holzstrukturen und die neuen Methoden der Roboterherstellung analysiert. Mit den gewonnenen Erkenntnissen wird es in Shanghai ein mittelgro��es Designprojekt im M50 Creativity Park mit Bauholz geben. Der Herstellungsprozess wird simuliert und parametrisch ausgelegt., This thesis researches the importance of digital fabrication and robotic timber constructions through analysing the historical background of digital fabrication. Different methods of digital fabrication will be researched and analysed and evaluated due to their advantages and disadvantages. Since a long-term CNC milling or cutting was already in use by industries and found its way to architecture. As well robotic arms were used in the automotive industry, but they were programmed to do one job for a lifetime. The computational design made it possible to use this flexible and precise tools to fabricate freeforms. Robotic arms have a high amount of advantages such as the flexibility in their movement, since they are not working in a cubic working space. As well the fact that they are not just attached to one tool, the end effector can be any kind of working tool. The robotic arm can execute every digital fabrication method started from additive method till to subtractive method, it depends on the end effector which is attached to the robot. The computational design is a huge milestone because of this, complex shapes and geometries can be fabricated with the robotic arm. The natural building product timber is a material which is not easy to manipulate for computational design but researchers around the globe invented different new methods for robotic fabrication which makes the computational design for timber products possible. In this research computational design of timber structures and the new methodologies of robotic fabrication will be analysed and with gained knowledge there will be a mid-scale design project in Shanghai at the M50 Creativity Park with construction timber. The fabrication process will be simulated and designed parametrically.

Robotic fabrication has expanded existing construction techniques, i.a. through enabling the assembly of bespoke structures made form discrete elements. This has substantially increased the design freedom, particularly through allowing designers to conceive and fabricate structures with complex geometries. However, this design freedom along with the intricacies of new fabrication methods has introduced new challenges regarding both digital design and materialisation. Robotic assembly, like other fabrication techniques, comes with its own constraints and limitations which are often difficult to intuitively describe. Similarly, bespoke geometries can result in complex geometric dependencies. This leads to a need for design methods that can control the resulting complexity and integrate it in the design process. The thesis is built around a cooperative robotic assembly method that utilises two robots to construct spatial structures made of steel bars. This expands existing assembly techniques by enabling the robots to change their role during the fabrication process while at the same time supporting each other. As a result, bespoke structures with complex geometries are built without the need of supporting or guiding structures. The proposed structures are based on a novel construction system that allows for a high level of differentiation. This is achieved particularly through a connection typology that allows to accommodate bars connecting at individual angles to each other. Furthermore, it presents a computational design strategy for geometrically, structurally and fabrication informed designs that is based on the assembly sequence and allows for a reciprocal information flow between design and fabrication. The goal of the thesis is to provide a set-up for designing and constructing spatial structures by connecting design and fabrication in one interlinked process. This is achieved through a combination of physical prototyping and digital design generation, by simultaneously developing the three main topics of the research: Constructive system, which includes geometric, structural and material systems, cooperative robotic assembly and computational design. Following hypothesis lays at the base of the thesis: Developing computational tools for simultaneously controlling the complexity of design, fabrication and structure leads to the mitigation of the limitations of prevailing spatial structures. This, in turn, allows architects and designers not only to fabricate complex designs but to envision new typologies of structures by gaining control and fully exploiting the design space resulting from fabrication, structure and geometry. Ultimately this leads to an expansion of the architectural design language.

This paper describes the design and fabrication of the ICD/ITKE Research Pavilion 2016/2017, the most recent in the series of experimental installations developed as an outcome of the design-and-build studio of the ITECH Masters Programme. The completed structure is a 12m long cantilevering lattice-composite shell that was wound in one piece by a multi-machine fabrication system using coreless filament winding. To realise such a structure through this fabrication process involved a negotiation between architectural design, structural requirements and fabrication constraints, details of which are found in this paper. Technical details of the multi-machine fabrication system were previously described in Felbrich et al.

Integrisanjem višerazličitih oblasti arhitekture, poznavanja svojstava materijala, fabrikacije, zajedno sa primenom digitalnih alata, postalo je moguće lakše I efikasnije fabrikovati složene arhitektonske forme. Primena penastog polistirena u arhitektonskoj fabrikaciji je pogodna zbog dobrih svojstava polistirena, prevashodno njegove lake obradivosti. Upotrebom zagrejane žice za obradu materijala I industrijskog robota kao mašine za fabrikaciju, u ovom istraživanju su pokazana tri projektantska scenarija, koja predlažu automatizovan proces fabrikovanja složenih arhitektonskih formi različitih veličina od penastog polistirena I primenom različitih strategija za fabrikaciju. Rezultati su predstavljeni u vidu fabrikovanih prototipova. Theintegrationofmultipledifferentfields-architecture,materialproperties,fabrication,combinedwiththeapplicationofdigitaltoolshasmadethefabricationofcomplexarchitecturalformseasyandefficient.Theapplicationoffoamedpolystyreneinarchitecturalfabricationissuitableduetothegoodpropertiesoffoamedpolystyrene,especiallytheeaseofmillingorcuttingthematerial.Inthisresearch,theapplicationofahot-wireasatoolandanindustrialrobotasafabricationmachinehasenabledthreedifferentdesignscenarios,whichsuggestanautomatedfabricationprocessforcomplexarchitecturalformsofdifferentsizes,madefrompolystyreneandbysuingdifferentfabricationstrategies.Theresultsarepresentsin a form of fabricated prototypes.

Standard industrialization and numeration models fail to translate the richnessand complexity of traditional crafts into the making of the architectural elements,which excludes them from the industry. This paper introduces a new way ofmodelling a complex craft fabrication method, namely the English Wheel, that isbased on the creation of a cyber-physical system. The cyber-physical systemconnects a robotic arm and an artificial neural network. The robot arm controlsthe movement of a metal sheet through the English wheel to achieve desiredgeometries according to toolpaths and predicted deformations specified by theneural network. The method is demonstrated through the making of 1:1 designprobes of doubly curved metal surfaces. Standard industrialization and numeration models fail to translate the richnessand complexity of traditional crafts into the making of the architectural elements,which excludes them from the industry. This paper introduces a new way ofmodelling a complex craft fabrication method, namely the English Wheel, that isbased on the creation of a cyber-physical system. The cyber-physical systemconnects a robotic arm and an artificial neural network. The robot arm controlsthe movement of a metal sheet through the English wheel to achieve desiredgeometries according to toolpaths and predicted deformations specified by theneural network. The method is demonstrated through the making of 1:1 designprobes of doubly curved metal surfaces.

Since the construction industry has begun deploying robotic technologies for digital fabrication processes, this direction has mostly been focused on integrating industrial-type robots into off-site prefabrication processes. By contrast, no enabling robotic technology exists today that allows robotic systems to be integrated into in situ construction processes right on the building site. This is mainly because in comparison with robotic prefabrication, robotic in situ fabrication faces fundamental technological challenges. First, buildings are large in scale. In contrast to prefabricating sub-assemblies of a building with stationary robotic systems off-site, in situ robotic systems must be able to fabricate large-scale assemblies at their final location. Second, building sites are poorly structured. As opposed to operate within structured factory conditions, in situ robotic systems must be able to accurately fabricate large-scale assemblies---irrespective of the uncertainties prevalent on-site. To firstly address the challenges imposed by the building scale, this thesis explores the mobile robotic fabrication of continuous and monolithic architectural-scale building components on-site. To further respond to the challenges imposed by the unstructured nature of building sites, and by the uncertainties related to the site, the fabrication, and the material, this thesis investigates the integration of robotic sensing solutions and the implementation of adaptive fabrication control techniques. For the purpose of validating the developed methods and strategies for robotic in situ fabrication, three subsequent case studies are implemented. Each of them utilises a different robotic set-up, once a stationary and twice a mobile, and applies a distinctive material system---that is, loam, bricks, and steel rebar. The research culminates in the third experiment, namely the mobile robotic fabrication of a doubly-curved steel rebar mesh for a reinforced concrete wall. This demonstration provides the unique opportunity to present robotic situ fabrication not only as a future vision, but applied in the context of a real construction project.

Combining computation and robotic fabrication and assembly creates rationale for new types of design and construction methods which are not bound to constraints of standardization and allows structurally informed differentiation. This research presents methods for designing and analysis of structural behavior for novel types of timber structures made of simple elements which are suitable for robotic assembly. Specifically, structures derived from Reciprocal Frames built with linear timber elements and connected with glued butt T-joints were of interest due to their applicability in integrated robotic fabrication and assembly processes. So far, little has been known about the complexity of their structural behavior and their potential for load-bearing application in construction. In response, this research established a corresponding structural analysis method for such structures and specifically the adhesive-based T-joint connections. Furthermore, this analysis method has been implemented as a computational tool integrated with algorithmic design modeling methods, which supports an architect or an engineer in exploring the potential designs based on immediate structural feedback. Finally, these tools were employed to conduct case studies which show that the structural behavior of the discussed structures is highly complex, and slight geometric modification allows to significantly improve the load-bearing capacity.

Fiber composite materials have tremendous potential in architectural applications due to their high strength to weight ratio and their ability to be formed into complex shapes. Novel fabrication processes can be based on the unique affordances and characteristics of fiber composites. Because these materials are lightweight and have high tensile strength, a radically different approach to fabrication becomes possible, which combines low-payload yet long-range machines, such as unmanned aerial vehicles (UAV), with strong, precise, yet limited reach, industrial robots. This collaborative concept enables a scalable fabrication setup for long span fiber composite construction. This paper describes the integrated design process and design development of a large scale cantilevering demonstrator, in which the fabrication setup, robotic constraints, material behavior, and structural performance were integrated in an iterative design process.

While fabrication is becoming a well-established field for architectural robotics, new possibilities for modelling and control situate feedback, modelling methods and adaptation as key concerns. In this paper we detail two methods for implementing adaptation, in the context of Robotic Incremental Sheet Forming (ISF) and exemplified in the fabrication of a bridge structure. The methods we describe compensate for springback and improve forming tolerance by using localised in process distance sensing to adapt tool-paths, and by using pre-process supervised machine learning to predict stringback and generate corrected fabrication models.