Physics-based Simulation of Agile Robotic Systems

Published

November 14, 2019

Author(s)

Pavlo Piliptchak, Murat Aksu, Frederick M. Proctor, John L. Michaloski

Abstract

Development and testing of industrial robot environments is hampered by the limited availability of hardware resources. Simulations provide a more accessible and readily modifiable alternative to physical testing, but also require careful design to ensure fidelity. This paper describes new progress in building such a simulation for NIST's Agility Performance of Robotics Systems (APRS) laboratory. This virtual environment is built and simulated using ROS and Gazebo respectively. The simulation also supports a Canonical Robot Command Language (CRCL) interface for added interoperability with physical lab components. The newest iteration of the APRS virtual environment reflects the addition of a conveyor and light curtain to the physical lab, and also features improved physics-based simulation. Despite physical accuracy being generally difficult to attain in robotics simulation, we demonstrate that the APRS virtual environment supports accurate simulation of kitting task actions such as grasping and object manipulation. Instrumental to achieving the current degree of accuracy was the development of several simulation components that provide general improvements to the simulation of conveyors, mimic joints, and contactrich scenarios. These components were developed and tested using elements of the APRS kitting task.

Conference Dates

November 11-14, 2019

Conference Location

Salt Lake City, UT

Conference Title

International Mechanical Engineering Congress & Exposition

Pub Type

Conferences

Download Paper

Local Download

Keywords

robots, agility, physics-based simulation, grasping, conveyor, kitting

Citation

Piliptchak, P. , Aksu, M. , Proctor, F. and Michaloski, J. (2019), Physics-based Simulation of Agile Robotic Systems, International Mechanical Engineering Congress & Exposition, Salt Lake City, UT, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=928067 (Accessed December 9, 2024)