Trimble, Shane (Queen's University Belfast), Naeem, Wasif (Queen's University of Belfast), McLoone, Seán (Queen's University Belfast)

Slip Signal Analysis on a Baxter Robot

Scheduled for presentation during the Regular Session "Robotics and Autonomous Vehicles" (WeA1Rob), Wednesday, August 21, 2019, 11:50−12:10,

5th IFAC International Conference on Intelligent Control and Automation Sciences, August 21-23, 2019, Queen’s University Belfast, Northern Ireland

This information is tentative and subject to change. Compiled on October 16, 2021

Abstract

The Baxter robot by ReThink Robotics is a dual arm collaborative platform, which can perform multiple tasks and is designed for use in an unstructured environment. Such robotic manipulators working in tandem have certain advantages over single arms when manipulating heavy or large, awkwardly shaped objects. With the attachment of two custom 3D-printed end-effectors, the robot gains the ability to push and hold objects of this kind.

This paper deals with the preliminary steps in creating a reliable controller feedback signal using the robot's on-board torque and positioning sensors. Determining the best joint signal and designing a suitable filter are two challenges that are addressed.

The paper concludes that the torque values from the centre wrist joint undergo the largest increase in value caused by vibrations during slip. A filtering array technique is also proposed, successfully removing unwanted noise from the torque signal, allowing slip events to be effectively isolated.