Regularized Deep Signed Distance Fields for Reactive Motion Generation

Puze Liu, Kuo Zhang, Davide Tateo, Snehal Jauhri, Jan Peters, and Georgia Chalvatzaki
Techinal University Darmstadt

Robots in Factory

Robot in Daily Life

• Safety

Robot in Daily Life

• Safety
• Unstructured Environments
• Delicate Manipulation
• Dynamical Human Behavior

Robot in Daily Life

• Safety
• Unstructured Environments
• Delicate Manipulation
• Dynamical Human Behavior

How can we achieve dexterous
manipulation tasks in dynamic
environment without collisions?

ReDSDF: A Regularized Deep Signed Distance Field

ReDSDF: A Regularized Deep Signed Distance Field

$$ d(\vx, \textcolor{darksalmon}{\vq}) = \left[1-\sigma_{\vtheta}(\vx, \textcolor{darksalmon}{\vq})\right]\textcolor{YellowGreen}{\underbrace{f_{\vtheta}(\vx, \vq)}_{\mathrm{NN}}} + \sigma_{\vtheta}(\vx, \textcolor{darksalmon}{\vq})\textcolor{orange}{\underbrace{\lVert \vx - \vx_c \rVert_2}_{\mathrm{Point\,Dist.}}}$$

ReDSDF: A Regularized Deep Signed Distance Field

$$ d(\vx, \textcolor{darksalmon}{\vq}) = \left[1-\sigma_{\vtheta}(\vx, \textcolor{darksalmon}{\vq})\right]\textcolor{YellowGreen}{\underbrace{f_{\vtheta}(\vx, \vq)}_{\mathrm{NN}}} + \sigma_{\vtheta}(\vx, \textcolor{darksalmon}{\vq})\textcolor{orange}{\underbrace{\lVert \vx - \vx_c \rVert_2}_{\mathrm{Point\,Dist.}}}$$

$ \sigma_{\vtheta}(\vx, \textcolor{darksalmon}{\vq}) = \sigmoid\left(\textcolor{OrangeRed}{\alpha_{\vtheta}}\left( \lVert\vx - \vx_c\rVert_2 - \textcolor{OrangeRed}{\rho_{\vtheta}} \right)\right) $

ReDSDF: Surface Reconstruction

Table

Shelf

ReDSDF: Surface Reconstruction

Table

Shelf

Tiago

Human

ReDSDF: Global Reconstruction

Reactive Motion Generation: Bi-Manipulation

Reactive Controller: Artifical Potential Field

Human Robot Interaction

Human Robot Interaction

Take Home Message

Regularized Deep
Signed Distance Fields Provides A Smooth, Fine-Grained Distance Function at Any Scale for Articulated Objects