On Effects of Continuous Event Approximations on Sensitivities and Optimization of Multibody Systems with Friction

EasyChair Preprint 13476

Abstract

This paper dives into the domains of sensitivity analysis and optimization of multibody systems with

frictional interactions. It explores the nuances associated with the approximation of discrete events in

friction model using high derivative continuous functions. Such approximations may work well when

it comes to computing dynamics, however the sensitivities may be very different. The study presents a

comparative analysis between a discrete event-based Coulomb friction model and a continuous approximation

Brown-McPhee friction model.

Keyphrases: Julia, Optimization, differential equations, estimation, events, hybrid systems, parameter, sensitivity

@booklet{EasyChair:13476,
  author    = {Adwait Verulkar and Adrian Sandu and Corina Sandu and Daniel Dopico},
  title     = {On Effects of Continuous Event Approximations on Sensitivities and Optimization of Multibody Systems with Friction},
  howpublished = {EasyChair Preprint 13476},
  year      = {EasyChair, 2024}}