Augmented Reality Driven Design and Fabrication of Shading Devices
Figure 1. Placement of a shadow area via the AR application
ABSTRACT
Sun exposure in urban environments regarding placemaking has become a larger area of research for city planners and architects. Placemaking is both a physical and psychological concept, as the severity of outdoor conditions and the perception of what conditions are considered normal can influence space usage. However, many urban spaces still need to be equipped to handle rising temperatures and broader placemaking needs. This paper proposes an augmented reality-driven design process for generating shading structures that can be utilized by non-experts via mobile devices. The generative algorithms streamlined the creation of design iterations and successfully utilized material simulation to develop shading devices based on looped surfaces. Furthermore, the integration of AR into the fabrication process simplifies construction, broadening accessibility and engagement for a diverse range of participants. Our research presents an accessible design approach and a fabrication methodology suitable for non-construction professionals, which can be applied in various contexts to provide shade with minimal infrastructural demands. We anticipate that this study will contribute to the expanding discourse on the application of AR in architectural design and fabrication, particularly concerning shading. We envisage that this accessible and engaging approach will promote wider community involvement in shaping urban spaces, offering viable solutions to rising temperatures in urban environments.
Keywords: Augmented Reality, Participatory Design, On-Site Design, Digital Fabrication, Shadow Analysis
1. INTRODUCTION
Urban public parks play an increasingly crucial role in offering recreational activities to diverse communities in urban locales. The design of these spaces needs to accommodate not just social activities, but also the experience of urban ecosystems, particularly in subtropical cities like San Antonio, where mitigating the impact of direct sun exposure is critical. Predictions suggest that by 2050, San Antonio will endure five months of temperatures exceeding 100.4°F (38°C), amplifying the existing severity of its climate (Boice and Garza 2021). The city council, recognizing these potential environmental challenges, has called for action (Office of Sustainability 2019). San Antonio studies emphasize the criticality of shade in improving thermal comfort in open-air spaces (Patel 2015), and recent case studies on shading structures underline the need to embed climate comfort into design considerations using computational tools (Ameijde et al. 2022).
In the context of architectural design, the use of digital design and fabrication can greatly influence the planning and creation of shading structures in urban parks. By harnessing computational design tools and adopting user-focused fabrication techniques, architects can make well-informed decisions to produce custom-made shading devices. Parametric design in architecture presents a unique opportunity to create numerous iterations of a single concept without the need for heavy human input (Figure 4-5). This approach can drastically speed up the design process by simplifying the process of drafting and managing design assets (Monedero 2000). A reconfiguration strategy has been proposed to embed interactive behavior in a shading canopy to ensure shadow in occupied spaces (Wood et al. 2019). However, the user doesn't have direct access to configuring or designing where and what kind of shadow is to be created.
The application of Augmented Reality (AR) – the overlay of digital information onto the real world – offers numerous innovative design opportunities. In architectural design and fabrication, AR finds utility in visualizing plans, monitoring progress, aiding workers, revealing hidden utilities, allowing remote experts to guide on-site personnel in real time, verifying code compliance of assembled structures, and enhancing the efficiency of construction equipment (Nassereddine, Veeramani, and Hanna 2022; Chi, Kang, and Wang 2013). Additionally, AR has been utilized in crafting complex architectural assemblies (Hahm 2019; Jahn, Newnham, and Berg 2022) and for various. However, its application remains limited, mostly due to unfamiliarity with the technology, environmental challenges at construction sites (Kassis et al. 2022), or implementation costs (Ramsey & Hongtao 2022), while industry trends suggest a growing adoption of AR (Noghabaei et al. 2020). The benefits of collaborative AR for interior design have been presented as early as 1995 (Ahlers et al. 1995). An extended reality collaborative design review process has been trialed with promising results, augmenting either a physical mockup or an existing environment in situ. This approach has demonstrated its potential to enhance inclusivity and active participation during design and planning processes (Gillespie et al. 2021).
Nevertheless, when it comes to non-experts' engagement in the design and fabrication process via AR technologies can be considered under-researched or not extensively explored. Additionally, the development of AR tools for user-guided on-site design and production of shading devices and the use of material behavior of bendable materials in this context presents unique contributions to this body of research.
In response to these challenges, we conducted a case study at [SITE] in Downtown San Antonio, an urban park grappling with heat-related issues. The choice of location was influenced by past design decisions that minimized shade, leading to lower-than-anticipated visitor engagement during spring and summer. The site's diverse programmatic demands require a response to various conditions. The client intends to install metal wires over the park, allowing for the flexible positioning of shading devices.
2. METHODOLOGY
Our research introduces an innovative method of creating user-defined shadow areas via an on-site AR interface on mobile handheld devices (Figure 1). 3D geometries are projected onto the planned wires above the site based on the identified shadows. These geometries are further refined through material simulations to define shadow density accurately. We also developed an AR fabrication tool, leveraging the properties of bendable materials, enabling on-site production of shading devices by participants with minimal construction knowledge. This tool provides detailed fabrication instructions for bending and securing a polycarbonate film into the desired shape. Our proposed methods are designed to heighten engagement and inclusion in shade creation, deploying human-centric AR tools during the design and fabrication stages. We validate this approach through the design and construction of a full-scale prototype.
This research followed the case study approach and was conducted in an academic setup at [UNIVERSITY] in collaboration with [CLIENT] and [COMPANY]. The design team consisted of a design-build course with fourteen students at the [UNIVERSITY], from January 18th to May 4th, 2022. We implemented and tested climate simulations, parametric design, and AR-assisted fabrication in this case study.
Figure 2. Mobile app concept diagram
DESIGN METHODS
We developed an augmented reality application designed to facilitate on-site positioning and evaluation of shading devices (see Figure 2). This tool enables users to specify the desired time for shadow creation by a projected shading device on-site. The projected shadow is then visualized at fifteen-minute intervals, distinguished by various shades of blue in the application.
The AR tool's geometric system is based on a bendable material, specifically chosen for its ease of fabrication. In our case, we used a thermoplastic film material (polycarbonate). The placement grid is circular in response to the material system of the bendable sheet. The size of these circles adapts based on the number of user-proposed shadows at different locations, with more proposals leading to larger circles. This feature provides users with insights into previous proposals, enabling them to either reinforce existing proposals or suggest other shadow locations.
This dynamic circular grid was realized using a particle-spring simulation, specifically Kangaroo for Grasshopper, in which the circle's radius reflects the frequency of user-placed points at the location. A threshold restricts the number and size of the circles to maintain a manageable visual representation.
Figure 3. The site constraints: the possible space (A), the adapted grid (B) for the placement of shading devices (C), and a layout option with shading devices placed between the wires with shadow from 15:00 - 16:00 on 15th of July (D)
In response to the site's specific requirements, we configured constraints for the application. These constraints included limiting possible locations for shading devices to align with a planned wire layout, which was dictated by two existing posts and a pergola (Figure 3). The tool prompts users to apply shadows to the site by marking points, which in turn generates a metaball polyline encircling these points (also shown in Figure 3). This metaball is then transformed into a mesh and projected to the wire height, roughly between four and five meters (approximately 14 to 16 feet).
This mesh is further translated into a pattern of bending and aligning curves that mirror the fabrication material's behavior. We utilized Rhino 7 software, combined with the Grasshopper and Kangaroo2 plugins, to generate these patterns. The boundary is then offset inwards and programmed to expand, provided it does not self-collide, thereby simulating the bending behavior of the polycarbonate.
Users have the flexibility to select from different pattern densities by altering the total length of the individual curve. The output is a dense pattern capable of providing the desired shadow. Generated patterns are user-controllable and can be modified by adjusting the input parameters, such as the offset distance and the rate of line growth.
Figure 4. The processes of pattern generation for the ribbon forms
The perforated designs resulting from this process aim to meet the programmatic needs while simultaneously preserving the vistas of the blue sky. The patterns have been calibrated to provide adequate shade during the noon and afternoon hours without causing excessive density that might obstruct sky views, thereby preventing a gloomy appearance of the park. The AR tool empowers users to focus shade provision over specific areas of interest or activity.
SHADOW ANALYSIS
A Universal Thermal Climate Index (UTCI) analysis was conducted and used to construct a thermal comfort map of the site. The UTCI calculation considers four main variables: dry bulb temperature, mean radiant temperature, relative humidity, and wind speed. Metrics of human comfort are estimated using universal data on human adaptive behavior, and other essential factors such as age, height, and weight are derived from population averages (Mackey et al. 2017). The relevant buildings, site vegetation, and shading devices were input as meshes into the Ladybug plugin in Grasshopper (Roudsari et al. 2013). A mesh was laid over the site with 2009 measuring points (Figure 5). For the measured period on July 15th from 11:00 AM to 2:00 PM, the site had an average air temperature of 32.2 °C and a mean radiant temperature of 54.6 °C based on the weather data from EnergyPlus for San Antonio. The measured UTCI temperatures were averaged and translated into a thermal comfort heat map.
Figure 5. A perspective showing the elevation of the shading devices above the ground and the mesh colored based on the heatmap.
AR-ASSISTED PROTOTYPING
The project's 3D forms were actualized by bending a continuous, thin film to conform to the generated patterns and forming joints at intersections of the film. These films, crafted from a lightweight, flexible, UV-resistant material – 0.02 inches (0.5 mm) thick and 48 inches (1219 mm) wide polycarbonate – made the bending process relatively uncomplicated. Moreover, the thermoplastic characteristics of the polycarbonate permitted it to be shaped without fracturing or needing a mold.
To support the construction of the prototypes, we used the Fologram app to establish an Augmented Reality (AR) aided construction environment. The AR model's primary function was to render the design visible in real-time as it was assembled and to offer a guide for the bending of the polycarbonate. This ensured that the prototype closely mirrored the intended design. In addition, the digital 3D model could be loaded onto AR-compatible devices, allowing it to be visualized during the construction process.
Construction tasks were divided into supervisory and building roles. Using an iPad handheld device, the supervisor monitored the overall model, ensuring that the prototype retained its correct form. Additionally, one individual donned an AR headset (Microsoft HoloLens2) to facilitate the construction process (see Figure 6).
Figure 6. 1:1 prototype assembly process. Supervisors on the left are comparing the built form to the AR model and inspecting it for any inconsistencies.
The builders, equipped with Microsoft HoloLens, guided the clamping and welding of the polycarbonate at a more local scale. A slider interface was implemented to create a more interactive process that allows the builder to alter the percentage of the AR model shown at a time (Figure 7). With this, the builders were able to visualize segments of the prototype without the overlay obstructing their vision.
Figure 7. The 1:1 augmented prototype assembly process, shows the geometrical overlay and the user interface.
Small-scale prototype iterations were constructed at various scales to test joint methods and the bending behavior. The smaller prototypes employed heat-based tools of varying temperatures, including hot air guns, impulse heat sealers, soldering irons, and welding guns. The types of welded joints tested include punctured fusion welds, continuous fusion welds, circular fusion welds, and W-shaped fusion welds (Figure 8). From the preliminary tests, intermittent W-shaped fusion welds were deemed the strongest of these options.
Figure 8. The W-shaped fusion welding tool and its application for the 1:1 prototype
For the 1:1 prototype, we used W-shaped fusion welds. We welded five points evenly spaced along the vertical axis at points where the polycarbonate film had to be joined. At most, a single joint connected five layers. The prototype joints were roughly aligned along a central axis to both strengthen it in the direction of the cables and to allow the prototypes to be compressed for storage. A two-story truss system was constructed on-site, and the 1:1 prototype was hung in the air for one week to test the feasibility of the joints. Several cameras were used to monitor the prototype for this duration, and its effects on on-site conditions were monitored.
3. RESULTS AND REFLECTIONS
This research presents AR-based on-site design and fabrication processes for shading devices based on computational design and shadow simulation guided by human decisions. Each step of the design process informs the next process. Collecting the necessary shadow analysis generated a design that efficiently responds to reduce direct sun exposure and improve thermal comfort.
AR BASED DESIGN
The AR design tool enabled non-experts to place shading areas as desired within the existing circular grid. Figure 9 shows a user during the process of using the cursor to select the shading areas while being able to get a view of the shading devices projected along the sun vector.
Figure 9. Stills from video capture during the placement of shaded areas via the AR application. The top row shows the placement on an iPad, and bottom row shows the placement as seen from a second device.
The scripts for layout and pattern generation streamlined the iterative phases of the design process. With it, a single designer was capable of producing and testing numerous designs within a limited time frame. In addition, the introduction of limited randomness contributed to the final design's aesthetic, which incorporated material properties and functionality. A final design was achieved through a material-based design system that utilized the polycarbonate's properties. In the context of shading structures, this process was used to find the pattern that maximizes shade while considering the material and fabrication constraints of the film material.
Figure 10. Examples of the script generating patterns of differing complexities and densities
SUN HOURS
Monochromatic shade models were constructed to calculate the concentration and total area of the shade cast by the structure (Figure 11). The location and area of the smallest and largest possible shadows cast were also determined. Control tests were run using solid blocks of equal dimensions for both models. Outputs were converted into percentages and used to gauge the effectiveness of the given forms.
Figure 11. Shadow analysis on July 21st from 11:00 AM to 3:00 PM for a flat surface without perforation (left), a solid geometry (middle), and the generated form (right).
A direct sun hour analysis was implemented, focusing on the immediate area surrounding a single, suspended structure (Figure 12). Using meteorological data from San Antonio, the hours of solar exposure on July 21st from 11:00 AM to 3:00 PM were calculated in 5 minutes increments.
Figure 12. Direct sun hour analysis on July 21st from 11:00 AM to 3:00 PM in 5 minutes increments.
Our measurements recorded both increases in shadow area and reductions in hours of direct sun exposure. As depicted in Figure 11, our pattern generates a maximum shadow of 5.07 m2 (cyan) and a minimum shadow of 2.8 m2 (pink). In contrast, the solid block generates a maximum shadow of 5.51 m2, while the flat surface generates a maximum shadow of 3.38 m2. Solar radiation analysis, shown in Figure 10, indicates that only minimal spots remain exposed to sunlight. The total area experiencing a reduction in solar exposure by more than an hour was 7.5 m2, and the overall area that observed reduced solar exposure amounted to 16 m2. Compared to the 3.4 m2 footprint of the polycarbonate form, the maximum area of solar protection is 4.7 times larger.
While the perforations do have some impact on the shadow area generated, the combination of orientation and depth in our design made this effect minimal. This is vividly demonstrated by the shading structure's smallest shadow; even though perforation allows some light penetration, the angle ensures this happens during times of the day with less cooling demand and in quantities too minimal to undermine its utility. The layout proposed in the UTCI analysis reduced the average temperature by 3-4° Celsius. The average perceived temperature in the analyzed area transitioned from 36.34 °C without intervention to 35.78 °C with the implementation of shading devices. In total, with the proposed twelve geometries, approximately 50 percent of the site's main lawn underwent temperature reductions significant enough to enhance thermal comfort. These results align with the temperature reductions offered by deciduous trees, as shown in Figure 13. Such results confirm the feasibility of larger-scale applications of our approach for AR-assisted shading design.
Figure 13. The UTCI heatmap for the site before intervention (left), including the shading devices (right).
PROTOTYPING
The results of the 1:1 prototype test underscored the concept's potential for creating comfortable pockets of shade. The entire construction process, greatly simplified by the use of AR-assisted design, was completed in less than a day by a team of just four builders. The AR technology served as an accessible blueprint, providing workers with clear assembly instructions. The segmentation of the AR model streamlined the information delivery to the builders, enhancing efficiency. Among the two devices used, the HoloLens goggles proved superior to the iPads, as they enabled workers to visualize the model while actively building.
The flexibility of this phase permitted modifications to the prototype during assembly. Builders could suggest changes to the pattern or the size of the curves within the constraints of ensuring future connections would remain secure and would not deviate substantially from the pre-established outline. The AR model could be swiftly updated to reflect any alterations and used to set a new endpoint if necessary.
During construction, the production of W-shaped fusion welds was both rapid and cost-effective. However, this process necessitated extra safety measures due to the production of fumes. While polycarbonate was used in this prototype, other more sustainable, eco-friendly materials could potentially be used in future iterations. For instance, bioplastics or recycled plastics could be considered, reducing the environmental footprint while maintaining the requisite flexibility and strength. Material selection should also consider potential outgassing or chemical release during the welding process, favoring materials that minimize these risks.
While the prototype's overall stability was satisfactory, we did note a single joint failure along the prototype's exterior. The timing and cause of this break are unclear, possibly due to mishandling during disassembly, but this incident underlines the need for stronger, reinforced connections in future models.
On-site monitoring confirmed the prototype's effectiveness in providing shade during the afternoon, particularly when the sunlight was directly overhead – the most crucial time for shade provision (Figure 14). Even when sunlight struck the prototype at an angle, the cast shadows were large enough to ensure thermal comfort. The prototype balanced its shading function with preserving views of the sky, avoiding the monolithic appearance of a solid shading surface, a testament to the potential of this approach.
Figure 14. The 1:1 prototype was installed with a provisional truss structure on site, with its shadow cast on the lawn (left) and the view through the perforated shading device (right).
4. CONCLUSION
In this paper, we have demonstrated a comprehensive approach to mitigating direct sun exposure via computational and on-site participatory design, shadow analysis, and digitally augmented fabrication, all enhanced by the power of augmented reality (AR). Our integrated design process caters to diverse programmatic needs and successfully improves thermal comfort while also addressing key placemaking necessities. Our tests confirm that our innovative structure effectively generates perforated shade, balancing sun protection with the maintenance of unobstructed views.
Our on-site design tools provide an in-depth comprehension of each shading intervention at the site scale, allowing users to focus on designing shadows rather than the shading device itself. Tailored computational design tools were developed explicitly for bending-active polycarbonate structures. Combined with the AR-assisted fabrication process, we offer a cost-effective solution that empowers a wide range of users to design and construct shading structures.
Our project expands the existing research striving to harness AR for design and fabrication, integrating material behavior through computational tools. The workflow we have developed can be applied to a multitude of projects across diverse sites, using 3D data of the surroundings to generate a site-specific shading solution. The simulation of material behavior within these computational analyses merits further exploration, with the potential to optimize the creation of perforated shading devices. Moreover, the incorporation of AR into earlier stages of the design process promises to democratize design and enable on-site construction by individuals without professional training.
This case study enriches the repertoire of available methods and tools for creating shade in urban parks experiencing extreme heat, like those in San Antonio. Our approach empowers a participatory design process and simplifies the construction process, making it accessible even to those without formal training. In summary, our project offers an innovative AR system to generate shade through gentle interventions, presenting a novel method for enhancing comfort in our urban landscapes.
REFERENCES
Ahlers, K. H., Kramer, A., Breen, D. E., Chevalier, P. ‐Y, Crampton, C., Rose, E., Tuceryan, M., Whitaker, R. T., & Greer, D. (1995). Distributed Augmented Reality for Collaborative Design Applications. Computer Graphics Forum. https://doi.org/10.1111/j.1467-8659.1995.cgf143_0003.x
Kassis, R. El, Ayer, S. K., Asmar, M. El, & Tang, P. (2022). Defining Factors That Support or Hinder Commercially Available Augmented Reality (AR) Devices for Construction Communication. In Computing in Civil Engineering 2021 (pp. 1302–1310). https://doi.org/10.1061/9780784483893.159
Ramsey, K., & Hongtao, D. (2022). The Perception of Virtual and Augmented Realities in the Construction Industry Based on Contractors? Perspectives in Washington State. In Computing in Civil Engineering 2021 (pp. 1311–1318). https://doi.org/doi:10.1061/9780784483893.160
Chi, Hung-Lin, Shih-Chung Kang, and Xiangyu Wang. 2013. "Research trends and opportunities of augmented reality applications in architecture, engineering, and construction." Automation in Construction, Volume 33: 116-122. https://doi.org/10.1016/j.autcon.2012.12.017.
Fologram. Accessed March 23rd, 2022. https://fologram.com/.
Gillespie, D., Qin, Z., & Aish, F. (2021). An Extended Reality Collaborative Design System: In-situ Design Reviews in Uncontrolled Environments. Association for Computer Aided Design in Architecture Annual Conference, ACADIA 2021.
Hahm, Soomeen. 2019. "Augmented Craftsmanship: Creating architectural design and construction workflow by augmenting human designers and builders." ACADIA, 448-456.
Jahn, Gwyllim, Cameron Newnham, Nick Berg. 2022. "Augmented Reality For Construction From Steam Bent Timber." CAADRIA.
Nassereddine, Hala, Dharmaraj Veeramani, and Awad S. Hanna. 2022. "Design, Development, and Validation of an Augmented Reality-Enabled Production Strategy Process." Frontiers in Built Environment, Volume 8. https://doi.org/10.3389/fbuil.2022.730098.
Noghabaei, Mojtaba, Arsalan Heydarian, Vahid Balali, and Kevin Han. 2020. "Trend Analysis on Adoption of Virtual and Augmented Reality in the Architecture, Engineering, and Construction Industry." Data, Volume 5 (1): 26. https://doi.org/10.3390/data5010026.
Song, Y., Koeck, R., & Luo, S. (2021). Review and analysis of augmented reality (AR) literature for digital fabrication in architecture. In Automation in Construction. https://doi.org/10.1016/j.autcon.2021.103762
Ameijde, Jeroen van, Chun Yu Ma, Garvin Goepel, Clive Kirsten, and Jeff Wong. 2022. "Data-driven placemaking: Public space canopy design through multi-objective optimisation considering shading, structural and social performance." Frontiers of Architectural Research, Volume 11 (2): 308-323. https://doi.org/10.1016/j.foar.2021.10.007.
Boice, Daniel, and Michelle E. Garza. 2021. “Urban Heat Island of San Antonio.” 43rd COSPAR Scientific Assembly. Abstract A0.4-0011-21 (oral), id.28.
Monedero, Javier. 2000. "Parametric design: a review and some experiences." Automation in Construction, Volume 9 (4): 369-377. https://doi.org/10.1016/S0926-5805(99)00020-5. Nassereddine, Hala, Awad S. Hanna, Dharmaraj Veeramani, and Wafik Lotfallah. 2022.
Office of Sustainability. 2019. "Vulnerability & Risk Assessment." The City of San Antonio.
Otani, Hidenori, and Jason K. W. Lee. 2022. "The use of sun-shade on safe heat exposure limit on a sunny summer day: a modelling study in Japan." International Journal of Biometeorology, Volume 66 (4): 731–740. https://doi.org/10.1007/s00484-021-02232-8.
Patel, Krishan. 2015. "Perceptions of Thermal Comfort: Landscape Design Attributes Based on The Shops At San Antonio." University of Texas at Arlington.
Wood, D., Yablonina, M., Aflalo, M., Chen, J., Tahanzadeh, B., & Menges, A. (2019). Cyber Physical Macro Material as a UAV [re]Configurable Architectural System. In Robotic Fabrication in Architecture, Art and Design 2018. https://doi.org/10.1007/978-3-319-92294-2_25