Deploying robots in real-world environments involves immense engineering complexity, potentially surpassing the resources required for autonomous vehicles due to the increased dimensionality and task variety. To maximize the chances of successful real-world deployment, finding a simple solution that minimizes engineering complexity at every level, from hardware to algorithm to operations, is crucial. In this dissertation, we consider a vision-based manipulation system that can be deployed in-the-wild when trained to imitate sufficient quantity and diversity of human demonstration data on the desired task. At deployment time, the robot is driven by a single diffusion-based visuomotor policy, with raw RGB images as input and robot end-effector pose as output. Compared to existing policy representations, Diffusion Policy handles multimodal action distributions gracefully, being scalable to high-dimensional action spaces and exhibiting impressive training stability. These properties allow a single software system to be used for multiple tasks, with data collected by multiple demonstrators, deployed to multiple robot embodiments, and without significant hyper-parameter tuning. We developed a Universal Manipulation Interface (UMI), a portable, low-cost, and information-rich data collection system to enable direct manipulation skill learning from in-the-wild human demonstrations. UMI provides an intuitive interface for non-expert users by using hand-held grippers with mounted GoPro cameras. Compared to existing robotic data collection systems, UMI enables robotic data collection without needing a robot, drastically reducing the engineering and operational complexity. Trained with UMI data, the resulting diffusion policies can be deployed across multiple robot platforms in unseen environments for novel objects and to complete dynamic, bimanual, precise, and long-horizon tasks. The Diffusion Policy and UMI combination provides a simple full-stack solution to many manipulation problems. The turn-around time of building a single-task manipulation system (such as object tossing and cloth folding) can be reduced from a few months to a few days.

Three-dimensional literature is a virtually non-existent or in any case very rare and emergent digital art form, defined by the author as a unit of text, which is not confined to the two-dimensional layout of print literature, but instead mediated across all three axes of a virtual space. In collaboration with two artists the author explores through a bodystorming workshop how writers and readers could create and experience three-dimensional literature in mixed reality, by using mobile devices that are equipped with motion sensors, which enable users to perform embodied interactions as an integral part of the literary experience.For documenting the workshop, the author used body-mounted action cameras in order to record the point-of-view of the participants. This choice turned out to generate promising knowledge on using point-of-view footage as an integral part of the methodological approach. The author has found that by engaging creatively with such footage, the designer gains a profound understanding and vivid memory of complex design activities.As the outcome the various design activities, the author developed a concept for an app called TEXTILE. It enables users to build three-dimensional texts by positioning words in a virtual bubble of space around the user and to share them, either on an online platform or at site-specific places. A key finding of this thesis is that the creation of three-dimensional literature on a platform such as TEXTILE is not just an act of writing – it is an act of sculpture and an act of social performance.