arXiv subjects

Maria Makarova

Publications and source records attributed to Maria Makarova.

6 recordsLinked to original sources

arXiv2025

DiffusionRL: Efficient Training of Diffusion Policies for Robotic Grasping Using RL-Adapted Large-Scale Datasets

Diffusion models have been successfully applied in areas such as image, video, and audio generation. Recent works show their promise for sequential decision-making and dexterous manipulation, leveraging their ability to model complex action distributions. However, challenges persist due to the data limitations and scenario-specific adaptation needs. In this paper, we address these challenges by proposing an optimized approach to training diffusion policies using large, pre-built datasets that are enhanced using Reinforcement Learning (RL). Our end-to-end pipeline leverages RL-based enhancement of the DexGraspNet dataset, lightweight diffusion policy training on a dexterous manipulation task for a five-fingered robotic hand, and a pose sampling algorithm for validation. The pipeline achieved a high success rate of 80% for three DexGraspNet objects. By eliminating manual data collection, our approach lowers barriers to adopting diffusion models in robotics, enhancing generalization and robustness for real-world applications.

cs.RO↗

arXiv2025

CONTHER: Context-Aware Reinforcement Learning for Robotic Manipulation with Sparse Rewards

This paper investigates whether sequential context improves goal-conditioned Reinforcement Learning in sparse-reward manipulation tasks. While Hindsight Experience Replay (HER) addresses reward sparsity through goal relabeling, its operation on isolated transitions limits its ability to capture temporal dependencies inherent in joint-space control. We hypothesize that incorporating motion history can enhance policy learning and introduce CONTHER, which integrates a Transformer-based architecture with a modified HER replay buffer. The Transformer encodes sequences of prior states and goals to provide temporal awareness, while the buffer populates experience with artificially successful trajectories. Two architectural variants are analyzed to examine how contextual information should be integrated. In simulated point-reaching tasks with a UR3 manipulator, CONTHER achieves a 38.46% higher average success rate compared to baselines and outperforms the strongest baseline by 28.21%, with faster convergence and more stable learning. The framework is further evaluated on three dynamic tasks requiring complex trajectory following and obstacle avoidance, where temporal context is critical. By operating directly on joint velocities, the approach provides a foundation for transfer to physical systems. The primary contribution is a systematic investigation into fusing sequential context with goal relabeling, offering insights into how temporal awareness benefits policy learning.

cs.RO↗

arXiv2024

LucidGrasp: Robotic Framework for Autonomous Manipulation of Laboratory Equipment with Different Degrees of Transparency via 6D Pose Estimation

Many modern robotic systems operate autonomously, however they often lack the ability to accurately analyze the environment and adapt to changing external conditions, while teleoperation systems often require special operator skills. In the field of laboratory automation, the number of automated processes is growing, however such systems are usually developed to perform specific tasks. In addition, many of the objects used in this field are transparent, making it difficult to analyze them using visual channels. The contributions of this work include the development of a robotic framework with autonomous mode for manipulating liquid-filled objects with different degrees of transparency in complex pose combinations. The conducted experiments demonstrated the robustness of the designed visual perception system to accurately estimate object poses for autonomous manipulation, and confirmed the performance of the algorithms in dexterous operations such as liquid dispensing. The proposed robotic framework can be applied for laboratory automation, since it allows solving the problem of performing non-trivial manipulation tasks with the analysis of object poses of varying degrees of transparency and liquid levels, requiring high accuracy and repeatability.

cs.RO↗

arXiv2010

Observation of Transparency of Erbium-doped Silicon nitride in photonic crystal nanobeam cavities

One-dimensional nanobeam photonic crystal cavities are fabricated in an Er-doped amorphous silicon nitride layer. Photoluminescence from the cavities around 1.54 um is studied at cryogenic and room temperatures at different optical pump powers. The resonators demonstrate Purcell enhanced absorption and emission rates, also confirmed by time-resolved measurements. Resonances exhibit linewidth narrowing with pump power, signifying absorption bleaching and the onset of stimulated emission in the material at both 5.5 K and room temperature. We estimate from the cavity linewidths that Er has been pumped to transparency at the cavity resonance wavelength.

physics.optics↗

arXiv2006

Silicon-based photonic crystal nanocavity light emitters

We have demonstrated an up to seven-fold enhancement of photoluminescence from silicon-rich silicon nitride film due to a single photonic crystal cavity. The enhancement is partially attributed to the Purcell effect. Purcell factor predicted by FDTD calculations is 32 for a linear three-defect cavity mode with computed quality factor of 332 and mode volume of 0.785 cubic wavelengths. Experimentally measured cavity quality (Q) factors vary in the range of 200 to 300, showing excellent agreement with calculations. The emission peak can be tuned to any wavelength in the 600 to 800 nm range.

physics.optics↗

arXiv2005

Two-dimensional porous silicon photonic crystal light emitters

We present the design, fabrication and preliminary experiments of two-dimensional photonic crystal cavities made in nanoporous silicon luminescent at 700-800 nm. Enhancement in photoluminescence extraction efficiency at the resonant wavelength is expected due to Purcell effect and directed radiation pattern defined by the cavity. Such cavities should also enhance nonlinearities exhibited by porous Si beyond what is observed in one-dimensional distributed Bragg reflection cavities due to their small mode volumes and modest quality factors. This design aligns itself well to integration with conventional silicon based electronics on a single chip.

physics.optics↗