D. Rahmati

Papers

Total Citations

H-Index

About

D. Rahmati’s research focuses on the intersection of reinforcement learning and robotic control, with a particular emphasis on precision and safety in human-robot interaction. Their most-cited work, “PPO and SAC Reinforcement Learning Based Reference Compensation” (2024, 2 citations), introduces a novel compensation method using proximal policy optimization (PPO) and soft actor-critic (SAC) algorithms to enhance the accuracy and smoothness of UR5e robot movements. This work is especially significant for rehabilitation applications, where robots must perform precise, tremor-free tracking of predefined trajectories to assist patients safely. By integrating reinforcement learning into reference compensation, Rahmati addresses critical challenges in adaptive control, enabling robots to correct deviations in real time without manual retuning. Though early in its citation impact, this contribution demonstrates a practical pathway toward more intelligent, responsive assistive robotics. Rahmati’s work bridges theoretical reinforcement learning advances with tangible robotic systems, offering a foundation for future developments in autonomous rehabilitation, industrial automation, and human-safe collaborative robots. Their research underscores a commitment to making robotic systems both more capable and more trustworthy in sensitive, real-world environments.