University of Twente Student Theses

Login

Optimization of product handling in the agricultural sector of an underactuated flexure-based gripper by using Bayesian estimation

Nijland, Lorenzo D. (2023) Optimization of product handling in the agricultural sector of an underactuated flexure-based gripper by using Bayesian estimation.

[img] PDF
2MB
Abstract:This research presents a method to optimally estimate the contact forces between an apple and an underactuated flexure-based gripper based on limited sensor information. Based on this estimate a controller maximizes the object acceleration while guaranteeing no-slip and no-crush conditions. Due to cleanability and robustness requirements present in the agricultural sector contact forces cannot directly be measured using force sensors on the gripper fingers, while it is also desirable to limit the number of internal sensors. To find the optimal control parameters an uncertainty parameterization of the apple is presented. In conjunction with a linear force approximation model a Kalman filter is used that allows to estimate the contact forces between the object and the gripper based on sensor information and control input. To find the minimal but still sufficient sensor types and combinations a comparison of single sensors and sensor combinations is made and they are assessed on performance. A test setup is built and used to validate and compare the results of the model and the real situation. In this particular apple usecase an increase in maximal allowable acceleration of about factor 2.4 in horizontal direction and 2.2 in vertical direction was found for an addition of 3 sensors compared to the no sensor approach.
Item Type:Essay (Master)
Faculty:ET: Engineering Technology
Subject:52 mechanical engineering
Programme:Mechanical Engineering MSc (60439)
Link to this item:https://purl.utwente.nl/essays/97702
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page