Adrian-Vasile Duka

Universitatea Petru Maior din Tîrgu Mureş

Papers

Total Citations

231

H-Index

About

Adrian-Vasile Duka is a robotics and intelligent systems researcher whose work centers on solving fundamental challenges in robotic control, kinematics, and autonomous navigation. His most significant contribution lies in applying machine learning techniques to the inverse kinematics problem — a notoriously complex mathematical challenge in robotics that involves determining the joint configurations needed to achieve a desired end-effector position. His 2014 paper, "Neural Network based Inverse Kinematics Solution for Trajectory Tracking of a Robotic Arm," has amassed an impressive 194 citations, establishing it as a key reference in the field and demonstrating the effectiveness of neural networks for trajectory planning in planar manipulators. Building on this foundation, Duka extended his research into neuro-fuzzy systems, publishing an ANFIS-based approach to the same inverse kinematics problem in 2015, further broadening the toolkit available to robotics engineers. Beyond manipulator control, his research also encompasses autonomous mobile robotics, including map-building algorithms that integrate odometry and image acquisition for environmental mapping. Collectively, Duka's work bridges theoretical computational intelligence and practical robotic implementation, making his publications valuable resources for researchers and students working at the intersection of machine learning and robotics.