Overview
Modern aerial robots are expected to operate in increasingly complex environments while interacting with humans.
Traditional mathematical tools often fail to handle complex sensor input, extreme agility control,
and indirect goal specifications.
They rely on explicit representations and goal specifications.
Moreover, explicit representations are often computationally expensive to obtain and reason over.
Therefore, implicit methods,
which leverage data and the underlying structures of perception, control, and planning,
are gaining increasing attention.
By embedding knowledge, constraints, and task specifications directly within learning frameworks,
implicit methods connect classical robotic paradigms with machine learning across multiple levels of abstraction.
Our Workshop will explore how implicit approaches, such as implicit representations (e.g., NeRF, Gaussian Splatting), implicit learning, and knowledge embedding, can be leveraged for aerial robots.
We aim to encourage a dynamic exchange of ideas between the audience and experts in control, planning, and perception, inspiring discussions on the role of implicit methods in the next generation of intelligent autonomy.
Sponsor
