Manipulator Performance Constraints for human-robot Cooperation

Description:

Use any local performance index of a robotic manipulator to constrain the robot above a certain threshold. This is particularly useful for singularity avoidance in physical human-robot interaction

A detailed description can be found in the following RCIM paper:

• Dimeas, Fotios, Vassilis C. Moulianitis, and Nikos Aspragathos. "Manipulator performance constraints in human-robot cooperation." Robotics and Computer-Integrated Manufacturing (2017).

and also in ICRA 2016 (early version):

• Dimeas, Fotios, et al. "Manipulator performance constraints in Cartesian admittance control for human-robot cooperation." Robotics and Automation (ICRA), 2016 IEEE International Conference on. IEEE, 2016.

A video demonstration:

This repository includes the Matlab code that was used in the ICRA 2016 paper and the extended method in C++ that is presented in the RCIM journal.

The C++ implementation includes performance constraints in both translational and rotational axes. It also includes 3 ways of calculating them:

• Serial: Each direction is calculated after the other. It might take a while
• Parallel: Each direction is calculated in its own thread worker. Function returns when all calculations are over. This usually works in less than 1ms. Better have a multi-thread CPU.
• Parallel non-blocking: Same as above but the function returns immediately. This can work in slow computers. WARNING! Be very careful when you use it, synchronization issues of the constraint forces may arise.

Usage:

C++

g++ demo.cpp performanceConstraints.cpp Jacobian.cpp -o demo -larmadillo -pthread -std=c++11

The Jacobian matrix that is provided in symbolic form is for the KUKA LWR 4+

Requirements:

• Armadillo library

Matlab

1. Set the desired parameters for the robot and the performance constraints
2. Run the simulation
3. View the simulated motion of the robot and the plots

Requirements:

• Robotics Toolbox for Matlab (http://www.petercorke.com/Robotics_Toolbox.html)

Notes:

• The simulation does not consider the joint limits so the robot might behave weird
• This code has been tested with Robotics Toolbox 9.10 and Matlab R2014a

Authors: Fotios Dimeas, Charalambos Papakonstantinou

Copyright 2015-2017 Fotios Dimeas