Correction notice (2026-08-03). This page was originally published in question-and-answer format containing dialogue attributed to Amy LaViers. No interview took place and that dialogue was constructed. It has been withdrawn and replaced with the source-verified profile below, which draws only on published work and public institutional sources. The Lab regrets the original publication and has since adopted an editorial rule prohibiting simulated interviews.

Amy LaViers is director of the Robotics, Automation, and Dance (RAD) Lab, a non-profit organisation based in Philadelphia working at the intersection of robotics and dance. She holds a PhD from Georgia Tech and is a Certified Movement Analyst through the Laban/Bartenieff Institute of Movement Studies — a combination that defines her contribution: bringing the analytical vocabulary of movement studies into engineering as a working method rather than a metaphor. Her work matters to somatic AI because it addresses movement quality as a designable property of machines, a dimension standard robotics metrics do not attempt to measure.

This is a source-verified profile, not an interview. It is assembled from published work and public institutional sources. No direct interview was conducted and no dialogue is reconstructed.

Position and background

LaViers directs the RAD Lab, which describes itself as an organisation for art-making, commercialisation, education, outreach and research at the intersection of robotics and dance. Founded in 2013, the lab reports having employed more than 40 people, produced three start-up companies, and received funding from the National Science Foundation, DARPA, industrial partners and university-based institutions. The lab was previously hosted at the University of Illinois before operating as a Philadelphia-based non-profit.

Her dual credentialing is central rather than incidental. Alongside doctoral training in electrical and computer engineering, she holds a Certified Movement Analyst qualification in the Laban/Bartenieff Movement System — the formal certification used by professional movement analysts. Published accounts of her work describe this movement training as having shaped her engineering research rather than sitting beside it.

The research approach

The RAD Lab's stated focus is the design of embodied abstractions and high-level control systems for robots with variable and perceptually meaningful motion. What distinguishes the method is the toolkit: alongside conventional dynamics, control theory and empirical measurement, the lab uses choreographic practice and movement taxonomies — specifically the Laban/Bartenieff Movement System — to study explicit, conscious strategies of embodiment and human movement creation.

The underlying claim is that movement quality is not an aesthetic residue left over after function is achieved. In applications involving physical human-robot interaction, how a robot moves is part of what it accomplishes. A movement analysis framework built over a century to describe qualitative distinctions in human movement provides vocabulary for specifying those properties, which conventional trajectory planning — which specifies position and velocity — does not.

Her published work appears in venues including Nature, Robotics and Autonomous Systems, IEEE Robotics and Automation Magazine, and with Oxford University Press, alongside presentations at Georgia Tech, Berkeley, Brown, Princeton and the DanceNOW Festival at Joe's Pub. She has also argued publicly, in writing aimed at engineers, that engineering requires qualitative methods — not as a supplement to quantitative work but as a necessary component of it.

Why this matters for somatic AI

LaViers' position identifies a gap that this platform has traced repeatedly through 2026: motion AI systems are overwhelmingly evaluated on whether a movement is physically valid and whether it looks plausible, with no third axis for whether it is qualitatively appropriate. The Lab has argued that this constitutes a missing third axis of movement evaluation.

The methodological proposal implicit in the RAD Lab's approach is that this gap is addressable through formal movement vocabulary. Laban-derived frameworks describe qualitative properties — the weight, timing, spatial and flow characteristics of a movement — in terms precise enough to function as design specifications. Whether such vocabularies can be operationalised into machine-learnable objectives remains open, and the Lab has set out the practical limits of using quality vocabulary with current generative systems in its guide to prompting AI motion generation with Laban effort terms.

Her work also belongs to a longer lineage in which computing is argued to require embodied and choreographic knowledge as a foundation rather than a consultation, developed across decades of interaction design research and continued in the Lab's profile of soma design and the trained perceiver.

Where to read further

  • The RAD Lab — the lab's own site is the authoritative source for current projects, personnel and funding.
  • LaViers, A., & Maguire, C. Making Meaning with Machines: Somatic Strategies, Choreographic Technologies, and Notational Abstractions through a Laban/Bartenieff Lens. MIT Press, 2023. Open access. The primary book-length statement of the method.
  • Google Scholar profile — for the full publication record, which is the appropriate source for her research claims in her own words.

References

LaViers, A., & Maguire, C. (2023). Making meaning with machines: Somatic strategies, choreographic technologies, and notational abstractions through a Laban/Bartenieff lens. MIT Press. https://mitpress.mit.edu/9780262546126/making-meaning-with-machines/

LaViers, A. (n.d.). Google Scholar profile. https://scholar.google.com/citations?user=qw0ukB4AAAAJ&hl=en

The RAD Lab — Robotics, Automation, and Dance. https://radlab.mechse.illinois.edu/

LaViers, A. (n.d.). Engineering needs qualitative methods. Medium. https://medium.com/@alaviers/robotics-automation-and-dance-d93589d60224