Kirill Safronov

Papers

Total Citations

H-Index

About

Kirill Safronov is a researcher at the intersection of robotics, computer vision, and biomechanics, with a focus on enabling safe and efficient human-robot interaction. His work addresses critical challenges in motion planning and object manipulation, particularly for large-scale and collaborative settings. Safronov’s most cited paper, “Neural Implicit Swept Volume Models for Fast Collision Detection” (2024, 7 citations), introduces a machine learning approach to accelerate collision detection—a traditionally time-consuming bottleneck in motion planning—by leveraging neural signed distance functions. This work has immediate implications for real-time robotic systems. In “Recognition and 6D Pose Estimation of Large-scale Objects using 3D Semi-Global Descriptors” (2019, 4 citations), he tackles the underexplored problem of recognizing partially visible, large objects critical for mobile robot navigation and manipulation. More recently, his 2025 study on “Integrative biomechanics of a human–robot carrying task” applies biomechanical analysis to design collaborative robots that assist individuals with sarcopenia, modeling assistance after human-human interaction. With a growing citation record and contributions spanning from foundational perception algorithms to applied assistive robotics, Safronov is shaping how robots perceive, plan, and physically cooperate with humans in complex environments.