Current Focus (Post-M.Sc. Graduation)

I am now independently exploring Vision-Language-Action (VLA) models for robotic manipulation, building on my expertise in robotics, SLAM, and multimodal AI. My recent work includes:

• Replicating SOTA VLA Systems: Implementing OpenVLA, Octo, and Pi0 in simulated environments (Gazebo/PyBullet) to benchmark their adaptability for dexterous grasping and language-guided tasks.

• Thesis-Driven Foundations: Leveraging my Master’s thesis (LiDAR-Camera-IMU SLAM Fusion for Industrial Digital Twins) to address sim-to-real challenges, particularly in sensor fusion and 3D reconstruction.

• Deep Learning Integration: Experimenting with PyTorch-based pipelines to combine diffusion models (DDPM/DDIM) with robotic policy learning for open-world object interaction.

I aim to bridge cutting-edge VLA research with real-world robotic applications. For collaboration or project inquiries, feel free to reach out.