IEEE Control Systems Society Presents Online Lecture: Force-Based Online Estimation and Adaptive Alignment for Robotic Peg-in-Hole Insertion

Thu, 26 Feb 2026

This presentation introduces a novel real-time control and estimation framework that enables precise robotic assembly without relying solely on vision systems. By reformulating peg-in-hole insertion as a state estimation problem, the approach uses force feedback to dynamically estimate hole location. A particle filter handles recursive belief updates, while Model Predictive Control (MPC) guides the robot toward informed probing points. Two path-planning strategies fixed and adaptive step sizes are introduced to balance speed and precision. Experimental validation on a UR10e robot arm demonstrates robust performance under varying misalignments, highlighting industrial relevance for high-precision automated assembly.

About the Speaker

Dr. Pramod Pashupathy

Postdoctoral Research Associate, Loughborough University