Robotic Building(s)

Abstract

Technological and conceptual advances in fields such as artificial intelligence, robotics, and material science have enabled robotic building to be in the last decade prototypically implemented. In this context, robotic building implies both physically built robotic environments and robotically supported building processes, whereas reconfigurable, robotic environments incorporating sensoractuator mechanisms that enable buildings to interact with their users and surroundings in real-time require design to production, assembly, and operation chains that may be (in part or as whole) implemented by robotic means. This paper presents and discusses research and experimental developments in robotic building implemented more recently at Hyperbody.

Structural Optimization for Materially Informed Design to Robotic Production Processes

“…Advancements in robotic building (Bier, 2014) as presented in this paper indicate that future building systems are customizable to fit demand driven production and D2RP processes enable material-and energy-efficient building. Such efficiency relies on materially, structurally and environmentally informed porosity, which at macro (building), meso (skin) and micro (material) scales, requires optimization of spatial configuration and material distribution with the aim to not only control mass-void ratios but also achieve an integral design, from overall building configuration to the achitectured material itself.…”

Environment Optimization for Multi-Agent Navigation

“…In this section, we analyze the convergence of the proposed method and provide an interpretation for the coordinated optimization scheme by exploring its relation with ordinary differential equations. We start by observing that solving the agent-environment co-optimization problem (1) with our method is equivalent to solving the following time-varying non-convex optimization problem The time-varying problem (12) changes with the navigation parameters θ a , which motivates the definition of a "time" variable α to capture the inherent variation. Specifically, θ a is updated by the actor-critic mechanism with policy gradient [cf.…”

“…In this context, we can represent θ (k) a as a function of the initial θ (0) a , the step-size ∆α and the number of iterations k, i.e., θ (k) a = Θ θ (0) a , α (k) where α (k) = k∆α is the step length. Define a continuous "time" variable α, which is instantiated as α (k) at iteration k. By combining this definition with problem (12), we can formulate a continuous time-varying problem as…”

“…where α ≥ 0 denotes the scale of "time" and θ o are the decision variables. Problem (13) is the continuous limit of problem (12), where the former reduces to the latter if particularizing the "time" variable α to discrete values {α (k) } k corresponding to different iterations. Because of the nonconvexity, we consider the first-order stationary condition as the performance metric for convergence analysis and define the local minimum trajectory of problem (13) as follows.…”

Towards Responsive Architecture that Mediates Place

“…The interrelated concepts of 'robotic buildings' [12], 'architectural robotics' [28] and 'robotic furniture' [86] hypothesise how the integration of robotic adaptation can nudge everyday human behaviours. We divide these research endeavours by whether this nudging is meant to be manifested statically, dynamically or spatially.…”