Abstract
Scientific knowledge, ideally neutral and impartial, is inevitably shaped by geographical, economic, and cultural contexts. This research contends that overcoming the constraints of human and economic scarcity, inertia, and limited funding access demands development of collaborative research networks. To this end, four university laboratories from Brazil and Chile - UFRJ, UBB, UTFSM and UAI - have united to advance robotics applied to architecture. The methodology begins with an analysis of Industry 4.0, Fab Lab implementations, and robotics in architecture in the region. They identify key research aspects by mapping each laboratory’s activities and technologies to pinpoint expertise and potential collaborative areas. The authors propose a summary table comparing the labs and a chronological overview to track regional robotics advancement. To raise awareness among peers, the initiative involves joint actions such as courses, workshops and technical visits. 1 Recognising the scarcity of robotic units and the inapplicability of procedures from wealthier contexts, the authors draw on their lab experience to propose guidelines for implementing robotics research units in academia within the region. This includes technological alternatives, installation considerations, and detailed configuration reviews. The deployment of additional robotic units is a means to foster collaboration and bolster the network. Anticipated outcomes encompass increased critical mass, collaborative research initiatives, faculty and student exchange across institutions, higher publication rates, knowledge acquisition, and improved access to global funding agencies. In summary, the Southern Creative Robotics network initiative aims to catalyse the creative integration of digital manufacturing and robotics technology within design in the Ibero-American context.
Southern Creative Robots
As described in previous surveys about Fab Labs in Latin America, such as Homo Faber,2–4 the Latin America region has different characteristics compared to other more industrialised regions in Northern Europe, USA and Japan. This research extends the area to the Ibero-American region, using the Global South concept. 5 This concept has a particular relation with the 4th industrial revolution in architecture addressed by Celani, 6 and by two IJAC special issues on the 4th revolution in Architecture.7,8 The Association Robots in Architecture has an International Map of Robots in the creative industry. 9 There are robotic fabrication laboratories in Chilean architecture and design faculties (UAI 2012, UTFSM 2014 and Bio-Bio 2019), and in Rio de Janeiro (2021) following the initial installations in Portugal (FAUP, DFL 2013) and Spain (IaaC 2011). An ongoing research by Pablo Herrera is currently preparing an updated map of robots operating in Latin America. 10 This article reports pioneering experiments and ongoing collaboration efforts of Southern Creative Robotics to provide a framework for implementing industrial robots creatively, in the Industry 4.0 context and the Fab Lab concept. The authors rely on their own experience, describing the lab activities concerned with robotics. Based on this experience, discuss robotics, tooling, actuators and sensors to integrate a robotic unit for creative manufacturing.
Industry 4.0, Fab Lab and robotics
In the fourth industrial revolution, information is omnipresent in the manufacturing process, in an unprecedented integration. 11 Cyber-physical systems merge physical, digital and biological information. Thus, compared to the machines of the third industrial revolution operated individually by humans, robots enable the integration of the production system as a whole. Robotics might collect data through sensors (force, temperature, proximity and humidity), the Internet of Things, barcodes, and increasingly resort to artificial intelligence and big-data that expand robotics capacity, from mechanical prostheses to agents, in an integrated network.
The application of robotics in Architecture has its own characteristics that are different from certain specific and mono-functional engineering contexts with pre-defined problems. Architecture as a discipline uses an open, exploratory process to discover and work on problems before defining them. A robot is an object of study used to rethink the entire manufacturing production chain. A project supported by design thinking uses empirical trial and error to explore problems that are unknown, concerning multiple domains. However, it is essential to have a vision of robotics focused on the construction industry, identifying the key areas to apply robotic processes fostering innovation. In short, this requires a holistic concept applied to construction, including materials, transformation processes and sustainability, supported by a Fab Lab-type structure. 12
Universities are increasingly adopting the FAB LAB concept: in this sense, even universities without a Digital Fabrication laboratory might consider transforming the model room, or the carpentry workshop into such a lab. For this transformation it is important to incorporate the maker culture and the “do it yourself” (DIY) approach. 13 The founder of the first Fab Lab at MIT in 2001 explains the fundamentals of overcoming a lack of knowledge, and the first question will probably be: “Why would the university need an industrial robot?” Pacini 14 presents a summary of the premises for introducing a portable unit to insert maker culture into traditional teaching. Similar concepts are also covered in texts from the PRONTO Móvel laboratory. 15
Documenting experiences
UTFSM – universidad técnica frederico Santa María, Chile
The experience at UFRJ, UTFSM, UAI and UBB point to the use of robotics for printing in mortar and various polymeric materials, ceramic materials and, lately, metal printing, as well as the development of stereotomy for wooden structures. The description of these pioneering experiments aims to help new laboratories regarding future activities in this large field with plenty to explore (Figure 1). UTFSM, credits to the authors’ laboratory.
The first experiments in Creative Robotics in Chile took place in the Architecture Department of the Federico Santa María Technical University, in the summer of 2013 16 at CIMA - Integrated Manufacturing and Automation Centre, of the Mechanical Engineering Department, Industrial Engineering and Electronic Engineering. The experiments trials tested the Grasshopper and KUKA PRC visual programming environment to control a KUKA KR125/2 industrial robot using the Windows 95 operating system with data transfer to the robot via a 3.5″ floppy disk. These tests aimed to assess the machine’s abilities – its tool and the 6 degrees of freedom of the industrial robot for subtractive manufacturing of geometries that required more than 3 degrees of freedom. Deformed trusses in thin MDF panels tested the generation of tool paths within a parametric design environment. 17 This required using Voronoi diagrams to machine with 4 degrees of freedom. In 2013, Robotics were introduced in the classroom through an architectural workshop led by visiting professor, Mauro Chiarella. 18 In 2014, the Department of Architecture acquired its own second-hand KUKA KR6 AGILUS robot, which, due to the lack of its own laboratory, was in CIMA until 2022.
Luis Felipe González Böhme (2014) received US$128,578 to research and develop a solution for the robotic reproduction of structural and ornamental elements of complex geometry, to restore and rehabilitate existing Chilean architectural heritage in wood. Luis Felipe González Böhme describes the research contributions. 19 One of the project’s external evaluators was Johannes Braumann, 20 the researcher who developed KUKA PRC. The R&D project allowed not only expanding the general knowledge, but also the tooling abilities. Ultimately, the project led to the creation of a new research line in the Department of Architecture called “Timber joinery robotics”. 21 In this new scenario, the application in construction of wooden structures without adding steel for the fittings was expanded to an incremental housing prototype 22 and to one for a bridge. 23 In 2016, UTFSM began teaching 4th and 5th semester architecture students to programme industrial robots in wood stereotomy applications. The student numbers have reached approximately 120 students per year. Research about carpentry with robotic assembly allows architects to expand the traditional carpentry’s range of wooden joints and assemblies (In the book “Uniones Carpintereas de Valparaíso” 24 ). This book has a detailed description of 22 constructive geometry classes, 5 assemblies and seams classes found in the load-bearing building structures between the mid-19th century and the beginning of the 20th century, in Valparaíso, Chile.
In 2018, thanks to collaboration with Universidad Bío-Bío, 25 the group acquired two collaborative robots UR5/CB3 for the Architecture Department. With this incorporation, in 2020, the group purchased a HUSKY A200 with a kit for a UR5/CB3, which, however, it was installed only in 2022 due to the COVID-19 pandemic. Between 2020 and 2022, Luis Felipe González Böhme obtained US$217,047 to research and develop a robotic carpentry assembly for the on-site prefabrication of flexible wooden housing solutions in the reconstruction processes of the Housing and Urban Development Service of the Valparaíso Region. The HUSKY A200 and UR5/CB3 integrated UGVs that allowed machining wood on the construction site, but the project’s main objective was to enable human-robot collaboration using machine vision with a ZIVID-TWO 3D camera. In 2021, in Valparaíso, the group founded the Robots in Architecture Area, 26 and, in 2022, the Robots in Architecture Laboratory 27 together with the Robotic Construction and Manufacturing Laboratory in Santiago. 28
UBB – universidad del bío-bío, Chile
In 2019, the University of Bío-Bío in Chile installed a KUKA 120 R2500 robot to experiment with 3D printing for construction. The robot was connected to a pump and a 120-L concrete mixer with an 8m hose clamped to the robot end-actuator.
29
A 10-L tube was attached with a piston at the end of the robot to print clay and other earth elements.
30
The group installed the robot on a 7 m linear rail, adjacent to a 6m industrial printer 12 m long by 4 m high, with a dedicated mixing pump, assembly tables, CNC cutting equipment, crushers and dry storage. The purpose was to build prototypes for construction elements based on concrete or wood (Figure 2). U. Bio-Bio, credits to the authors’ laboratory.
For printing, the group developed a cementitious mixture with support from local industries for an urban furniture repertoire, 31 artificial reef samples and house wall prototypes. 32 The process developed proposals in Revit and Rhinoceros programmed for the toolpath in KUKA PRC. It was important to consider the use of the site for wet works, with mobile platforms, large parts storage, and the participation of two or three operators to control the work of the robot and the pump/mixer. In addition, the preparation of study models (mock-ups) with plastic and clay printing helped to develop details and design processes. 33 Initially, UBB developed these activities in a joint project with the Architecture Department of the Universidad Técnica Federico Santa María, which had initiated the use of robots in this field in Chile. Other tasks tested with this robot were lifting blocks, and the design development of a CNC wooden gripper with double scissors to grasp objects by approximation, eliminating the need for an electric motor. This device can lift wooden and concrete blocks, and assemble a construction system avoiding human effort with precise but flexible execution. 29 Nowadays, the lab researches BIM-models linked to printing building parts, seismic resistance, and develops a constructive system for local housing.
UFRJ - universidade federal Rio de Janeiro, LAMO-PROURB, Brazil
LAMO-PROURB – Laboratório de Modelos 3D e Fabricação Digital is a laboratory at the Universidade Federal do Rio de Janeiro founded in 2013. LAMO has a design culture grounded in full-scale building construction. This “constructivist design culture” strives to incorporate computational design processes. These design processes and digital fabrication require validation in practice. In other words, LAMO believes in an integrated production chain that includes conception, fabrication, robotics and construction. Construction is the ultimate design proof. Some examples of this constructivist culture are the wiki-house project built in 201434–36 and the Tornado Ruled Surface Pavilion in 2017.36,37 To develop full-scale constructions incorporating computational thinking demands vision. The lab developed this along the way, incorporating it in digital design fabrication processes (subtractive, additive and transformative) (Figure 3). LAMO, credits to the authors’ laboratory.
The Faculty of Architecture and Urbanism (FAU-UFRJ) is not yet digitally oriented; LAMO assured machinery, such as laser cutters and 3D printing, available for the students and researchers. The arrival of these tools required introducing the students to programming, first in workshops and then progressively in elective courses. Computational processes are required to include the fabrication tools in the design processes. The lab’s technological update requires more funding, which, in the Brazilian context, is demanding. This obstacle, on the other hand, has led the lab to develop the maker and DIY culture, building 3D printing machines, and interactive control mechanisms with arduíno, including small prototypes and robots as in the Workshops Abrigos Sensíveis in 201438,39 and Defying Gravity in 2015.40,41
Our involvement with robotics increased with José Pedro Sousa’s invitation in 2014 to visit the DFL Laboratory in Porto (location of the first robotic unit in architecture in Portugal 42 ), together with an invitation to share our lab research. LAMO applied for funding for a robotic unit in 2015, in a LAMO/NANO/COPPE proposal for the Rio State Research Foundation, FAPERJ. The lab submitted a proposal for an industrial robot unit with a group of interactive robots approved by the funding agency, but, in the end, it did not receive the investment. Finally, in 2019, the lab received financial support from a public innovation agency, FINEP, to buy and install an industrial robot, a KUKA KR 120-270r. The robot arrived in Rio at the beginning of 2021. FAU-UFRJ was the first Public Architecture University in Brazil to receive such an industrial robot. Ironically, to install it, the lab had already developed 10 projects for 10 different locations, and the expectation is to conclude the installation in 2024. The lab preparation had the support of William Barbosa, who had installed a robotic arm in the PUC-Rio engineering department to design and implement the robotic unit. 43
The lab’s first experience in robotics was the “Building Proto-Ecologies, building with robots” event organised by NANO (EBA-FAU) in a partnership with LAMO and the Bartlett University, at the Museu do Amanhã (Museum of Tomorrow) in Rio, in 2016. 44 Our partners from Bartlett organised a design workshop with a robot provided by a company in São Paulo. There was a problem with the instructor, and Gonçalo Castro Henriques led the participants to design and assemble a customised brick wall, taking as reference the work of Gramazio and Kolher. This involved developing a quick visual code to assemble a task related to others used in digital manufacture, such as in CNC, to create tool paths to place bricks considering the robot’s singularities. Meanwhile, the instructor from London arrived in Rio and set up the available robot to assemble the proposals developed by the workshop participants. 45
Through the years the lab research has advanced on additive processes, developing its own manufacturing machines for additive paste printing.46,47 Parallel with these actions in digital fabrication initiatives, LAMO has promoted computational design through teaching visual and textual programming for a new generation of computational literati. 48 Over the years, LAMO has developed partnerships to leverage research. An example is the research partnerships, in existence for some years, with Rodrigo Alvarado U. Bío-Bío, Chile, 1 and others, partnerships such as with Paulo Fonseca, who conducted applied research in robotics, 49 Regiane Pupo (Pronto3D) and Gabriela Celani (LAPAC/PLASMA), just to name a few. Collaboration enabled the research as a whole to advance with more breadth and speed to compensate for the increased difficulty in acquiring investment, thus helping to develop and share scientific knowledge in applied research.
UAI - universidad adolfo ibañez, Santiago, Chile
Universidad Adolfo Ibañez and the Design Lab School of Design founded a Fab Lab in October 2011.
50
The lab’s mission was to enable cutting-edge practical technological training in digital manufacturing and advanced manufacture for undergraduate and postgraduate students, as well as to be a reference for researchers, developing knowledge and transferring technology to society. Originally, it was located in a small space near Adolfo Ibanez University, on the President Errazuriz Campus, in Providencia, before moving to the new building on the Peñalolén Campus, in Santiago. During the first months, machines arrived and were installed, such as Epilog laser cutters and engravers, a large bed Multicam Water Jet V-Series KMT NEO40i, a Stratasys Dimension 3D 2000 ES printer and several other machines. In 2014, UAI opened a second Fab Lab on the Viña del Mar Campus. UAI research is resumed in Figure 4. UAI, credits to the authors’ laboratory.
April 2012 marked the arrival of the first robot, a KUKA KR180 R2500 with a KUKA KRC2 controller and an HSC spindle mounted on a static foundation. These settings enabled performance of the first tests and training, before moving to the Peñalolén Campus in May 2013. UAI organised the Fab Lab 51 in three spaces: a small lab space shared with the Bioengineering Department, UAI Engineering School, to place biomaterial equipment, bioreactors, a digital microscope, a lyophilizer, a small universal loading-cell testing equipment, and several growing chambers. The second lab was the prototyping Fab Lab (Lab 105) that concentrated small-scale equipment for this purpose, namely: vacuum forming, laser cutting and small 3D printers, including the Stratasys Dimension 1200 ABS 3D printer with dissoluble support, an array of desktop PLA 3D printers, the circuit maker Roland Modela MDX20 desktop mill, and Roland Modela GX24 cutter. Together with the welder and electronics, there were workbenches to design and assemble microcontrollers for custom applications, such as end controllers for the Kuka robotic arm. The School dedicated a third lab space (D004) to large-scale industrial processes. This lab included a CNC Lathe CK 6140s, a vertical panel saw and several electronic tools for wood work; a robotic unit with KUKA KR180 on a 5m linear rail, a rotatory base, an automatic tool-changer, end-tools such as the grippers, spindle, plus other tools on a secondary tool-changer.
The final location of Robotic and Industrial Processes Fab Lab has a Kuka robot on a 5m linear rail, with increased reach to LWH: 2.8 × 5.0 × 3.6 m. In 2014, the authors installed an additional rotatory base with a 200 kg payload to machine complex geometries with accurate synchronization, to define the robot-starting angle and avoid self-collision events in programming routines. In 2022, a fourth laboratory (D001) was prepared for materials and product research, biomaterials growth, and for testing universal materials.
UAI Methods: this approach was to test robotics for digital production and to automate fabrication of complex structures in multimaterial systems using a set of robotic manufacturing techniques. The techniques repertoire ranged from formative, subtractive and additive manufacturing methods to automatic complex product parts assembly. Hence, there has been gradual investment in equipment to accumulate progressively a technologies and techniques repertoire, not only to introduce students to automated multiaxial fabrication using robotics, but also to expand research capabilities in industrial products and architectural elements in full-scale experimental design.
The initial explorations went beyond cutting and milling polymeric and wooden materials with the HSC spindle to implementing new tools for cutting, milling, 3D printing and scanning with the robotic unit. Some tools were custom made at the Fab Lab, increasing the range of materials and industrial applications. In 2014, UAI taught the first Digital Fabrication class both undergraduate and graduate students from the Design Engineering programme about subtractive manufacturing techniques. The techniques used the spindle to cut and mill polymers and wood, and the foam cutter to produce large complex formwork (see Figure 4, bottom of first image), and assembly of complex double-curved masonry walls assembled by grippers (see Figure 4, top of first image).
Robotic labs (case studies).
Group developments by year.
Robotic labs initiatives timetables
Cooperation developments
The first collaborative initiative Figure 5 in robotics in architecture in the southern hemisphere, started with Mauro Chiarella visit (FADU-UNL), invited to jointly organise a workshop with Luis Felipe González Böhmeat UTFSM, in Valparaiso, Chile, in 2014
18
). Through the years, there were collaborative initiatives among Argentina, Chile, Brazil, Portugal and Spain regarding design and digital fabrication.
1
In 2020, there was a joint initiative from four universities: UFRJ (Brazil), UBB (Chile), FADU-UNL (Argentina) and UCR (Costa Rica), in a jointly organised workshop held in Rio de Janeiro. This seminar and workshop had a scarce budget and machinery, and focused on sharing research experience, and designing gridshell structures in small-scale prototypes (2021). Although the Rio event did not use robotics manufacture, it constituted an important impulse for the future agenda. From this event came the idea to write a paper for SIGraDi on how to install a robotic unit in our region,
56
and the idea to foster future exchange experience in developing robotics. Southern creative robotics 2023. Map of the partner universities. The field visit started in Rio de Janeiro (UFRJ, left upper row of 1st photo), them Valparaíso (UTFSM, 2nd photo), then Santiago (UTFSM, 3rd photo and UAI, bottom row of 4th photo) and then Concepción (UBB, 5th and 6th photos).
Field visit Chile 2023
At the end of 2022, the group planned a scholar exchange in the form of a 12-day field visit (2nd - 14th June 2023), ranging across 7,500 Km, starting from Rio de Janeiro, passing through three cities and four universities Figure 6: UTFSM in Valparaíso, UTFSM and UAI in Santiago and UBB in Concepción. This was the first event for Southern Creative Robotics intended to strengthen collective intelligence and practical knowledge by understanding the specificities, research activities, infrastructures and technologies of each group, using a more holistic knowledge exchange concerning techniques and researches to leverage the group as a whole. Southern creative robotics 2023. Activities by row: UTFSM Santiago; UTFSM Valparaíso, UTFSM Santiago; UAI Santiago; UBB Concepción.
The field visit’s first stop was at Universidad Frederico Santa Maria UTFSM Valparaíso for a two-day event. On the first day, the hosts Luis Felipe González Böhme– Director of the department of Architecture and founder of the local Robotics Group - together with the local researchers, Eduardo Valenzuela and Matias Correa, presented the facilities on the Valparaíso Campus, facing the Pacific Ocean shore. They presented the design department, the robotic facilities, and the on-going research on robotic fabrication. This visit helped to understand their local university culture and way of working. For the second day, the hosts organised a group of lectures about robotics. With the arrival of Francisco Quitral from the research group and Professor at the UTFSM in Santiago, along with 50 students from the city. There were another 50 from Valparaíso, thus composing an audience of 100 students, accompanied by six professors. 57 The UTFSM students had attended parametric design and digital fabrication classes with the research group. Professor Luis Felipe González Böhme hosted the sessions. He himself gave a lecture (robotic framing carpentry), and lectures by Gonçalo Castro Henriques (computational network experiences), and Eduardo Valenzuela (robotic manufacturing assisted by augmented reality). The lectures stimulated the participants, giving rise to discussion about technology and robotics in design. UTFSM in Valparaíso has a strong technical tradition, as the design department is included in the engineering school. The first initiatives in robotics mentioned involved successfully “hijacking” the robots from their repetitive tasks in mechanical engineering, with the help of foreign professors, Chiarella and Calvo, the aim of which was to explore other design possibilities. This initiative led to research connected with the application of robotics to expand timber construction in the heritage city of Valparaiso. The research results enabled financial support to assemble a design-dedicated unit in Santiago do Chile, with its own industrial robot. Nevertheless, the university in Valparaíso kept a compact robotic unit - Laboratorio de Construcción y Manufactura Robotizada - focused on robot mobility and virtual reality to foster on-site timber construction.
The second stop was at Campus San Joaquin, UTFSM Santiago do Chile. Rodrigo Alvarado joined the group, coming from Valparaíso with local professors. The activities included a visit to the lab facilities and a workshop on robotics. The lab is a large industrial shed; the main space had a 6-ton metallic structure designed to hold a rail for the robot (KUKA Quantec KR 210 R3100-2 C), hung upside-down to increase the degree of freedom; under the structure, there was a mechanical pit with a removable metal grate. The main space had subspaces attached dedicated to fabrication such as CNC, laser-cutting and manual cutting machines. Adjacent there were also two large rooms for student classes/activities, and ones on an upper floor with research facilities. The neighbouring shed research is about material development in civil engineering. The visit started with participation in a Robotic Workshop – “Carpintería de Armar Robotizada” – conducted by Francisco Quitral. The students from Valparaíso and Santiago had a project to prepare robotic timber assemblies for fabrication. It was a good opportunity to test robotic fabrication in real time. The students tested their fabrication for mortar-tenon assembly; in the lab, a metal grid protected the robot from overenthusiastic students eager to see the materialization of their design, the robot being the attention centre. Another day, there was a visit to Studio GT2P Great things to People with Sebastian Rozas (Universidad Chile), who introduced us to the rich manufacturing experiments blending tradition with digital, fuzzy techniques and local craft. 58 Finally, we learned wood carpentry in Museo Taller, dedicated to traditional wooden tools, techniques and culture, in order to learn how to innovate through the rich heritage of Chilean carpentry.
The tour continued, visiting the UAI Universidad Adolfo Ibañez at Peñalolén, in the Chile Metropolitan region, at the Andes foothills. The campus, designed by Architect José Cruz Ovalle, spreads across the mountains in an organic pattern. The UAI is an Engineering University, with a course on design engineering. It hosted the UAI fab-lab, one of the first fab-labs in the country. The Director of the Design School and design Lab, Mario Vergara received us and introduced us to the facilities. The fab-lab uses two spaces, on two overlapping floors, densely occupied by machinery. On the upper floor, there was a 3D printing farm with different additive materials and technologies, in addition to laser cutting machines, vacuum forming, and small-scale robots (KUKA Agilus KR4-600) for everyday use. The lower floor housed machines for full-scale project development, such as a plasma cutter, CNC cutter, and a robotic arm (KUKA KR180 R2500). Finally, the University Dean, Prof. Sergio Araya, Professor at UAI, researcher and visiting scientist at MIT, received us and presented the UAI University. Together we discussed their experience and a framework for future collaboration.
The final stop was at UBB Universidad del Bío-Bío, in Concepción, a college-town in south-central Chile. This university is a reference in the region, and its compact campus hosts diverse research areas that establish inter-departmental co-operation research. The visit started with the UBB Professor and Researcher, Paula Ulloa showing us the campus facilities. Then we moved to the INES Innovation Centre, a creative building designed by the architect, Pezo von Ellrichshausen. It features a central void in exposed red-pigmented concrete. The reception was by Claudia Muñoz Sanguinetti, Professor of the Department of Construction Sciences and Research Director of the CITEC-UBB Construction Technologies Research Centre. The CITEC had already built three experimental buildings on the UBB campus. Rodrigo Alvaro met us there to show the CITEC shed occupied by the School of Engineering and Construction that housed the fabrication lab jointly used with the Architecture department. Together we visited the concrete probes, and watched material tests with the team using the KUKA 120 R2500. In the lab, we could witness the collaboration between departments in the same shed where they tested concrete materials, conducted 3D printing, welding, and connected with BIM management. Next on the itinerary came a visit to Bancapar (FADU-UNL/UBB) and the wall Muro Pixel, together with 3D printed urban furniture. In the afternoon, there was a meeting with masters’ students from the Magister “Latinoamericano” in Architecture, directed by Rodrigo Alvarado, who introduced the lecture, “Trayectorias computacionales, experiencias en red” by Gonçalo Castro Henriques, UFRJ LAMO-PROURB. At the end of the day, there was a visit with Professor Maurício Vargas to PymeLab Tower, an experimental wooden tower in CLT, besides universities and research labs; part of the field visit also addressed the Chilean cities and the rich architecture and construction that have their own blend of tradition and technology (Figure 7). Southern creative robotics 2023. Field visit summary: gathering with gathering with González, Quitral, Henriques and Garcia (authors of the article together with Andres Passaro and Sergio Araya) at UTFSM Santiago; UTFSM Valparaíso conference (1st photo in the upper row), UTFSM Santiago workshop on robotics (2nd and 3rd photos in the upper row); at UAI, a visit to the lab facilities on the Peñalolén campus; UBB lecture post-graduation Latin-American, visit to UBB lab, and presentation of the robotic experience in research (bottom row).
A Robotic manufacturing unit
Robotic unit setup
Robotic units have different setups according to the application type, their research and practical needs. Below is a description of robotic unit types for academic research in the university environment. The setup types presented are Robotic Manufacture Unit; Robotic Container Unit; Robotic Prototyping Unit; Robotic in Situ Manufacturing Unit; Robotic Modelling Unit; Robotic Production Unit (Figures 8). Some of these setups are complementary. This classification is based on a book by Gramazio and Kolher.
59
As these architects based their investigations on their experience, do not considered these examples in a reductive way, as technology evolves with mobility increase, the incorporation of artificial intelligence, and new human-machine collaboration interfaces. The book, “Towards a Robot Architecture”
60
explores these subjects in greater detail. Despite these limitations, the specific units’ descriptions are useful to request support from founding agencies for generic units. From top to bottom, left to right: robotic manufacturing unit; robotic container unit; robotic prototyping unit; robotic manufacturing in situ; robotic modelling unit; robotic production unit.
Robotic manufacturing unit
A unit is generally composed of a medium or large robotic arm installed on 6m rails, or on a fixed base. The arm reach is 1.70 m to 3.0 m. Equipped with an automatic spinning table, designated as a positioner that is normally associated with an automatic changing tool. Generally, it is installed in interior areas (rooms with dimensions LWH: 10 m × 7 m × 4.40 m), and uses the robot brands: KUKA, ABB or Staubli adapted from Gramazio and Kohler. 61
Robotic container unit
It is composed of a transportable container with a medium or large robotic arm inside, the arm reach being 1.70 m to 3.0 m. The container (LWH: 6.0 m × 2.2 m × 2.4 m) allows the top lid and front door to be opened. Mounted on a rail the length of the container for external projects, in more isolated places. The robots are normally KUKA and ABB.
Robotic prototyping unit
This unit is generally composed of at least three small robots, in series, with an arm reach of up to 0.9 m. Linear positioning on a robot table (LW: 5.4 m × 1.8 m) allows serial tasks on an assembly line. Generally, the make, Universal Robots is used.
Robotic manufacturing in Situ
This is composed of a medium or large robotic arm with a reach of 1.7 m to 3.5 m. Robot installed on a mobile caterpillar-type vehicle. Free movement allows tasks that require extensive mobility. It uses robots of the KUKA and ABB makes, but there is also a mobile platform from Universal Robots. One of Kuka’s most recent lines is adaptable to more complex and variable situations, its mobility and use of small robots are advantages in the teaching environment.
Robotic modelling unit
This is generally composed of makes of small or medium-sized robots with low load capacity and arms ranging from 0.9 m to 1.6 m. The robots are positioned on rotatable bases (LWH: 1.5 m × 1.5 m × 3.8 m) without rails. Used to model light 3D elements such as Styrofoam and plastic.
Robotic production unit
This comprises a set of robots housed in a shed (LWH: 45 m × 15 m × 8 m), generally with robots hanging from overhead cranes, others on the floor on mobile rails or in separate positions. Generally, the operating radius of the robots intercepts to facilitate serial production. Usually, the KUKA ABB and Saubli makes are used.
Costs estimation
Robotic manufacturing unit.
Values: Another perverse situation when purchasing relates to understanding is what constitutes a minimal functional robot. The Industrial Robot is just part of the “Unit’s” “gear”, so we recommend an in-depth checklist with an expert on the subject. If any part is missing, it will delay the unit’s operation.
Maintenance process
When purchasing, another important matter, not always taken into account, is the maintenance routines for the equipment. In Brazil, recently from FINEP, the value of the equipment must be at least US$100,000. Therefore, when purchasing an arm and positioning table, or a rail and accessories, it is important to find out the value of these notices in the state/federal agencies, and check that the total value of the invoice exceeds this value by some margin.
Software demands
Embedded software, in addition to being expensive, is generally unnecessary. Currently, the Association for Robots in Architecture has developed a plug-in called PRC (Parametric Robots Control) for Grasshopper, specifically for KUKA robots, and there are more laboratories building this plug-ins, which are generally free, or with affordable annual fees. This plug-ins enable generation, operation and simulation of G-Code from Grasshopper, embedded within Rhinoceros, priced at around US$200, in sharp contrast to amounts between US$7000 and US$10,000 that sellers charge for specific tasks.
Tools and accessories
Robot manufacturing companies offer some accessories for automation based on the specificities of production (Industry 3.0), and manufacturing labs generally design and make their own accessories, or purchase separate parts and assemble them. In this case, it is worth studying the accessories needed for the robot in depth and requesting an amount approximately half the value of the robotic arm for this purpose: Spindle, Electric Finger and Pneumatic Finger, suction cup, welder, extruder for medium and large 3D printing ports, mortar projector, etc.
Installation/safety considerations
It is also worth setting aside another amount equal to half the cost of the arm for installation. The most expensive installation is electrical, but it might be necessary to consider water, sewage (to perform 3D printing in mortar a sandbox is important), pneumatic installation, dust extractor, compressed air, and an appropriate base, as, generally, to fix the arm, it is essential to drive stakes into the ground to withstand the different bending moments. Guaranteeing the security of the location usually demands a cage or sensors that stop the arm’s operation when someone approaches. There are several standards, and sellers are keen to indicate all the possibilities. Nevertheless, for those who have attended any workshop addressing responsive matters, via Arduino, with sensors, you can reduce these values from a DIY. The ideal situation is to have space on the ground floor for the robot to operate freely, with an expanded rotation of 370° and a passage clearance of at least 1m, space to move different worktables, as well as space to store immediate production, and possibly tool change equipment. Site construction is a complex activity, and the design with the help of someone experienced.
Industrial robot and digital manufacturing unit
Basically, a digital fabrication laboratory revolves around several aspects, and the laboratories described here, according to the Homo Faber research, 2 are labs aimed at building 1/1 pavilions, and we share our perspective in accordance with this experience.
As mentioned, there is a preference for the KUKA make over others because it has developed open source plug-ins and KUKA PRC. Today, there are plug-ins for other makes, but there is still a desert of news. For manufacturing tasks, aiming at 1/1 scale for pavilions, the ideal is to purchase a robust robot of the Quantec line. The nomenclature type is KR 120 R2700, where: KR = KUKA Roboter, 120 = payload or workload, which, means that a work tool weighing 120 Kg can be placed, although in the calculation it is necessary to consider friction, etc. R2700 means the equipment’s 2.7 m working radius. As mentioned, the cage makes up a square of 5.40 m on a side. We place a tool in a use area of 6m, which, including the traffic safety area, gives a total square of 8 m × 8 m.
Conclusions about the robotic framework
This article presents a repertoire of technological alternatives and practical considerations for implementing a robotic unit for teaching and research in architecture, thus promoting more than a strict technical vision. Rather, there is a comprehensive vision that includes the different criteria summarised below. The main industrial robot manufacturers: KUKA, ABB, UR, FANUC produce similar robots in terms of hardware, designed as manipulators for repetitive industrial tasks, in work cells and structured environments, such as in the automotive industry. According to our experience and that of partners from other countries, the main differentiating factor when choosing a robot is the possibility of using visual programming, which is provided by KUKA PRC and HAL. It makes it friendlier for practicing architects, designers and artists. Originally, the creators of industrial robots did not conceive them as machining tools for agile CAD-CAE-CAM manufacturing processes, as it was necessary for the Fab Labs. Learning from other machines such as CNC, laser cutters and 3D printers, although linked to the maker culture, does not directly prepare for the use of an industrial robot, given the increase in complexity and its specificities. The implementation of robotic equipment for teaching architecture faces the dilemma between installing a larger, more expensive industrial robot for full-scale construction activities, but on fewer occasions, and the alternative of several smaller collaborative robots, which allow development of more activities, involving more students and researchers. Regardless of the choice, the best preparation for the creative use of robots is to teach parametric design, and visual and textual programming. Learning to programme is vital in several areas, and this can change the project paradigm. 48
To organise a creative robotics laboratory with an industrial robot requires space, infrastructure, training and development of specific tools. There is a recursive idea to develop machines (and codes) that can develop new machines or tools, and learn in the process in a supervised environment, envisioning the use of Artificial Intelligence. The industrial robot will certainly play a central role in organising new spaces and the tasks, and, in the case of expanding existing laboratories, should integrate not only the physical aspects, but also the existing capabilities and demands. Defining the tools involves experimenting with different materials, tasks and accessories, which must be associated with project development ideas. The robot, in addition to mechanical tasks, can also develop intelligent activities, which require specialised knowledge framed in the design context. To implement a robotic unit, the concepts of Industry 4.0 are relevant to encourage integrated development of other computations and our own robotic abilities.
Discussion: On the advantages of network collaboration
The partners’ joint work has resulted in recommendations, validated by their experience, for a robotic framework adapted to the southern hemisphere context, expanding and strengthening the network. There are also other interesting outcomes, such as the partners’ awareness of what the other members have done, are doing, and intend to do, as a platform for establishing possible synergies and collaborations, for example, in relation to additive, subtractive and transformative processes and technologies, considering developing customised robotic end-effectors and tools. A networked strategy like this can avoid material, equipment and research duplication, as there is no room for redundancy if you have scarcity, promoting interdisciplinary and international cooperation. The possibility to learn from the partnership teaching contexts, their universities and specific audiences, and each specific relation with local industry can promote a quicker maturation of local technology.
A relevant impact of such a network is also the multidisciplinary collaboration and the expansion of professional profiles in teaching, normally considered solely as development in strict disciplinary areas and conventional industry tasks. While the use of robotic systems, especially in contexts of resource scarcity and handcraft development, motivates self-learning, communication between different specialties, direct experimentation, and the adoption of adapting strategies to modify, upcycle or re-invent existing artefacts and processes. This approach permeates disciplinary boundaries and encourages new generations to associate and work collaboratively, mixing manual and automated labour in the search for global solutions. Integrating and reviewing local experiences of small industries or operators provide situated knowledge, inspire and spur new applications, as they converge in a documented catalogue of experiences and external references that stimulate further development and quality perspectives inside traditional industries.
Within the described partnership framework, considering robotic equipment as an exploration partner, articulates human capabilities and technical possibilities with circulation of people and knowledge, in an atmosphere of co-operation and trust, enabling collective efforts to advance further, deeper and faster. Are there only advantages? The answer is “No”, as there are major challenges now and in the near future, such as whether to feed or co-ordinate the network effort from a distance. The Southern Creative Robotics project is a seed, and future results should confirm its importance.
Footnotes
Acknowledgments
LAMO-PROURB thanks all collaborators who participated in the research and the support FAPERJ, CNPq and MCTI/FINEP/CT-INFRA-04/2018. The Bio-Bio University had the support of the National Commission for Scientific and Technological Research of Chile, Scientific and Technological Development Support Fund FONDECYT 1221730. The Federico Santa María de Valparaíso Technical University had the support FONDEF ID20I10262.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Fondo Nacional de Desarrollo Científico y Tecnológico, Chile (FONDECYT 1221730); Fondo de Fomento al Desarrollo Científico y Tecnológico, Chile (FONDEF ID20I10262) and Minist´erio da Ciˆencia, Tecnologia e Inovação, Brasil (MCTI/FINEP/CT-INFRA 04/2018 Agreement 01.19.0184.0).
