Diego Romeres
Mitsubishi Electric Research Laboratories
80 Papers
81 Citations
Diego Romeres is an academic researcher from Mitsubishi Electric Research Laboratories. The author has contributed to research in topics: Computer science & Reinforcement learning. The author has an hindex of 9, co-authored 46 publications. Previous affiliations of Diego Romeres include University of California, Santa Barbara & University of California, Berkeley.
Chat about Author
Papers
Tactile-RL for Insertion: Generalization to Objects of Unknown Geometry
Siyuan Dong,Devesh K. Jha,Diego Romeres,Sangwoon Kim,Daniel Nikovski,Alberto Rodriguez +5 more
- 30 May 2021
TL;DR: In this article, a tactile feedback insertion policy is proposed to align the object and environment with a tactile-based feedback insertion strategy, and the insertion process is modeled as an episodic policy that iterates between insertion attempts followed by pose corrections.
130
Online semi-parametric learning for inverse dynamics modeling
Diego Romeres,Mattia Zorzi,Raffaello Camoriano,Alessandro Chiuso +3 more
- 01 Dec 2016
TL;DR: In this article, a semi-parametric algorithm for online learning of a robot inverse dynamics model is presented, which combines the strength of the parametric and nonparametric modeling.
59
Sim-to-Real Transfer Learning using Robustified Controllers in Robotic Tasks involving Complex Dynamics
Jeroen van Baar,Alan Sullivan,Radu Cordorel,Devesh K. Jha,Diego Romeres,Daniel Nikovski +5 more
- 20 May 2019
TL;DR: In this article, the authors propose to learn robustified controllers in simulation, which are learned by exploiting the ability to change simulation parameters (both appearance and dynamics) for successive training episodes.
46
Derivative-Free Online Learning of Inverse Dynamics Models
TL;DR: In this article, a derivative-free (DF) framework is proposed for inverse dynamics modeling in robotics, which does not require this preprocessing step and outperforms existing methodologies.
Trajectory Optimization for Manipulation of Deformable Objects: Assembly of Belt Drive Units
Shiyu Jin,Diego Romeres,Arvind Ragunathan,Devesh K. Jha,Masayoshi Tomizuka +4 more
- 30 May 2021
TL;DR: In this article, the authors formulate the belt drive unit assembly task as a trajectory optimization problem with complementarity constraints to avoid explicitly imposing contact mode sequences, and solve the problem as a mathematical program with complementary constraints (MPCC) to obtain feasible and efficient assembly trajectories.
38