Ph.D. Student at the Montanuniversität Leoben

Short bio: Mr. Vedant Dave started at CPS in August 2021.
He received his Master degree in Automation and Robotics from Technische Universität Dortmund in 2021 with the study focus on Robotics and Artificial Intelligence. His thesis was entitled “Model-agnostic Reinforcement Learning Solution for Autonomous Programming of Robotic Motion”, which took place at at Mercedes-Benz AG. In the thesis, he implemented Reinforcement learning for the motion planning of manipulators in complex environments. Before that, he did his Research internship at Bosch Center for Artificial Intelligence, where he worked on Probabilistic Movement Primitives.
Research Interests
- Movement Primitives
- Probabilistic Learning
- Deep Reinforcement Learning
- Machine Learning
Contact & Quick Links
M.Sc. Vedant Dave
Doctoral Student supervised by Univ.-Prof. Dr. Elmar Rueckert since August 2021.
Montanuniversität Leoben
Franz-Josef-Straße 18,
8700 Leoben, Austria
Phone: +43 3842 402 – 1901 (Sekretariat CPS)
Email: vedant.dave@unileoben.ac.at
Web Work: CPS-Page
Web Private: https://www.linkedin.com/in/vedant-dave-095629178/
Chat: WEBEX
CV of M.Sc. Vedant Dave
DBLP
Frontiers Network
GitHub
Google Citations
LinkedIn
ORCID
Research Gate
Publications
2021 |
|
Rozo*, Leonel; Dave*, Vedant Orientation Probabilistic Movement Primitives on Riemannian Manifolds Conference Conference on Robot Learning, vol. 5, 2021. @conference{Rozo&Dave*2021, Learning complex robot motions necessarily demands to have models that are able to encode and retrieve full-pose trajectories when tasks are defined in operational spaces. Probabilistic movement primitives (ProMPs) stand out as a principled approach that models trajectory distributions learned from demonstrations. ProMPs allow for trajectory modulation and blending to achieve better generalization to novel situations. However, when ProMPs are employed in operational space, their original formulation does not directly apply to full-pose movements including rotational trajectories described by quaternions. This paper proposes a Riemannian formulation of ProMPs that enables encoding and retrieving of quaternion trajectories. Our method builds on Riemannian manifold theory, and exploits multilinear geodesic regression for estimating the ProMPs parameters. This novel approach makes ProMPs a suitable model for learning complex full-pose robot motion patterns. Riemannian ProMPs are tested on toy examples to illustrate their workflow, and on real learning-from-demonstration experiments. | ![]() |