RSS

Robotics: Science and Systems paper. Two-column, experimental robotics with real hardware focus.

Category

Conference

License

Free to use (MIT)

File

rss/main.tex

main.texRead-only preview
\documentclass[conference]{IEEEtran}
\IEEEoverridecommandlockouts
\usepackage{cite}
\usepackage{amsmath,amssymb}
\usepackage{graphicx}
\usepackage{booktabs}
\usepackage{hyperref}

\begin{document}

\title{Continuous-Time Trajectory Optimization\\for Soft Manipulators}

\author{\IEEEauthorblockN{First Last and Jane Doe}
\IEEEauthorblockA{\textit{Department of Robotics, University of Example}\\
\{you, jane\}@example.com}}
\maketitle

\begin{abstract}
Soft manipulators promise safer, more compliant physical interaction,
but their rich dynamics make planning costly. We propose a
continuous-time trajectory optimization framework that models the soft
manipulator as a piecewise-polynomial field and uses Lie-theoretic
integration. Our method plans 22$\times$ faster than prior FEM-based
methods while matching their accuracy on three benchmark soft arms.
\end{abstract}

\section{Introduction}
Soft manipulators offer compliance benefits but are difficult to
control. Existing planners either approximate them as rigid or accept
very slow FEM-based integration.

\section{Related Work}
Piecewise constant curvature, Cosserat rod models, learned dynamics.

\section{Method}
We represent the manipulator backbone via piecewise-polynomial Lie
algebra elements. The optimization problem becomes:
\begin{equation}
\min_{\boldsymbol\xi} \int_0^T L(\boldsymbol\xi(t), \dot{\boldsymbol\xi}(t)) \, dt
\end{equation}
subject to boundary and contact constraints. A 4th-order geometric
integrator preserves group structure.

\section{Experiments}
\begin{table}[t]
\centering\small
\begin{tabular}{lcc}
\toprule
Method & Planning (s) & Tracking err (mm) \\
\midrule
FEM-based      & 184 & 4.2 \\
PCC approx     & 3.1 & 11.8 \\
\textbf{Ours}  & \textbf{8.3} & \textbf{4.6} \\
\bottomrule
\end{tabular}
\end{table}

\section{Real-Robot Experiments}
Deployed on a 12-segment pneumatic arm, our planner achieved 94\%
success on pick-and-place across 50 trials.

\section{Conclusion}
Continuous-time geometric optimization unlocks fast and accurate
planning for soft manipulators.

\bibliographystyle{IEEEtran}
\bibliography{refs}
\end{document}
Bibby Mascot

PDF Preview

Create an account to compile and preview

RSS LaTeX Template | Free Download & Preview - Bibby