Skip to content

Set of controllers for managing human-robot-environment interactions

License

Notifications You must be signed in to change notification settings

ze3000A/interaction_controllers

 
 

Repository files navigation

interaction_controllers

Licence Build

This package contains the implementation of the following controllers for ros2_control that are used for physical human-robot-environment interaction control.

Impedance Controller

Impedance Control is a widely used interaction control method, particularly efficient for human-robot interactions. It consists in imposing an impedance model for the relationship between the manipulator and the environment.

Admittance Controller

Whereas IC requires the system to be controlled in effort, Admittance Control is a control strategy that allows to impose an impedance model generating position commands.

Model Predictive Impedance Controller (MPIC)

Model Predictive Impedance Control is a control strategy based on Model Predictive Control (MPC) that allows compliant behavior when interacting with an environment, while respecting practical robotic constraints. A detailed description of the controller algorithm can be found in the following article:

@inproceedings{ Bednarczyk_MPIC_2020,
 	title={Model Predictive Impedance Control},
  	author={Bednarczyk, Maciej and Omran, Hassan and Bayle, Bernard},
	booktitle={2020 IEEE International Conference on Robotics and Automation (ICRA)},
	place={Paris, France},
  	year={2020},
  	month={May},
 	pages={4702-4708}
}

Variable Impedance Controller With Passivity Guarantees (VIC)

The Variable Impedance Control With Passivity Guarantees is a control strategy that allows online modification of impedance parameters while guaranteeing the overall system passivity. A detailed description of the controller algorithm can be found in the following article:

@inproceedings{ Bednarczyk_VIC_2020,
	title={Passivity Filter for Variable Impedance Control},
	author={Bednarczyk, Maciej and Omran, Hassan and Bayle, Bernard},
	booktitle={2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
	place={Las Vegas, USA},
	year={2020},
	month={Oct},
	pages={7159-7164}
}

Contacts

icube

ICube Laboratory, University of Strasbourg, France

Maciej Bednarczyk: m.bednarczyk@unistra.fr, @github: mcbed

About

Set of controllers for managing human-robot-environment interactions

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • C++ 82.6%
  • C 8.9%
  • CMake 8.5%