Video

 
 

Contact

 
Antonio Franchi
Phone: +49 7071 601 218
Fax: +49 7071 601 616
antonio.franchi[at]tuebingen.mpg.de

 
to Personal Site
   

Haptic Remote Control and Enhanced Interactivity

The focus of this project is to enhance the interaction between a human operator and a multi-robot system or a single robot autonomously performing a task. This is relevant for tasks taking place in remote or dangerous environments.
The overall system, as depicted in Fig. 1, is composed of several parts:
•    one or multiple mobile robots (slave system) that perform a task in a remote environment. The slave structure includes
•    onboard computational units (GNU-Linux PC Board)
•    onboard sensors (Inertial Measurement Unit, camera, etc.)
•    inter-robots communication infrastructure
•    one or multiple human operators (master system) that command some desired motion primitives or aspects of the task to the robots. The master control is realized via several interfaces
•    one or more haptic devices (actuated devices) through which the operator gives commands and receives a force feedback relative to the execution of the task
•    monitoring devices that provide additional information on the slave system (video stream, internal state of the robots)
•    other input devices
•    a master/slave communication channel (there can be delays, packet losses)
Our approach provides an extension of traditional bilateral (force-feedback) teleoperation systems to the control of multi-robot formations and to the trajectory planning.

An illustrative example of the bilateral teleoperation of a robot formation is depicted in Fig. 2. In this case we have that
•    the slave is a formation of robots with an assigned desired formation up to a scale factor.
•    the master controls size, position and rotation along a vertical axis of the whole formation via two haptic devices (using 5 DOF)
•    the force feedback is informative of the mismatch between the commanded motion (to the formation) and the executed motion.

The application to trajectory planning, depicted in Fig. 3, considers the following scenario
•    the slave system is a trajectory parameterized by some geometric parameters (shape variables)
•    the master commands via a haptic device some desired behavior to the trajectory (translation, resizing of the shape, change in the circulation speed)
•    the robot traveling the trajectory autonomous corrects it to avoid collisions and guarantee feasibility
•    the force-feedback given to the operator is based on the mismatch between the commanded trajectory and the corrected one (it’s a feedback on future planned motion)

Selected Publications

Franchi A Person, Masone C Person and Robuffo Giordano P Person (October-2012) A Synergetic High-level/Reactive Planning Framework with Application to Human-Assisted Navigation 2012 IEEE IROS Workshop on Robot Motion Planning: Online, Reactive, and in Real-time, 15-20.
Masone C Person, Franchi A Person, Bülthoff HH Person and Robuffo Giordano R Person (October-2012) Interactive Planning of Persistent Trajectories for Human-Assisted Navigation of Mobile Robots IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2012), IEEE, Piscataway, NJ, USA, 2641-2648.
pdf

Export as:
BibTeX, XML, Pubman, Edoc
Last updated: Thursday, 02.05.2013