M2CURL: Sample-Efficient Multimodal Reinforcement Learning via Self-Supervised Representation Learning for Robotic Manipulation
Fotios Lygerakis ,
Vedant Dave ,
Elmar Rueckert
arXiv 2024
project page
/
paper
M2CURL builds on MViTac to improves RL by efficiently integrating visual and tactile representations. It accelerates learning in downstream manipulation tasks.
Your browser does not support the video tag.
Multimodal Visual-Tactile Representation Learning through Self-Supervised Contrastive Pre-Training
Vedant Dave* ,
Fotios Lygerakis* ,
Elmar Rueckert
ICRA 2024
project page
/
code
/
paper
MViTac integrates vision and tactile modalities using contrastive learning, focusing on their inter and intra-modality relationships.
Your browser does not support the video tag.
Can we infer the full-arm manipulation skills from tactile targets?
Vedant Dave ,
Elmar Rueckert
Workshop on Advances in Close Proximity Human-Robot Collaboration, Humanoids 2022
paper
Using a multi-image diffusion model as a regularizer lets you recover high-quality radiance fields from just a handful of images.
Your browser does not support the video tag.
Predicting full-arm grasping motions from anticipated tactile responses
Vedant Dave ,
Elmar Rueckert
Humanoids 2022 (Oral Presentation)
paper
TacProMPs learns and predicts complex arm movements based on tactile responses, particularly for manipulating non-uniform objects, demonstrating adaptability diverse grasping scenarios.
Your browser does not support the video tag.
Orientation Probabilistic Movement Primitives on Riemannian Manifolds
Leonel Rozo* ,
Vedant Dave*
CoRL 2021
paper
Using a multi-image diffusion model as a regularizer lets you recover high-quality radiance fields from just a handful of images.