PhD Position

Title: development and control of tensegrity mechanisms for interventional radiology applications

Workplace: LIRMM, Montpellier, France
Thesis director: Philippe Poignet, Professeur, University of Montpellier
 Supervisor: Salih Abdelaziz, Assistant Professor, University of Montpellier
Starting date: from October 2017 (3 years)
Keywords: interventional radiology, robotic assistance, design, modelling, identification and control

1. Context

Tensegrity mechanisms, a particular class of prestressed systems, receive nowadays increasing interest in the robotic community. They are mechanical assembly of a set of compressed bars and a set of tensioned elements that can be either cables or springs. The actuation is performed either through the cables or the bars. They exhibit interesting properties, such as a large workspace and a good stiffness to weight ratio. Moreover, they offer the possibility to be remotely actuated using long cables. This possibility is of a great interest to enhance the mechanism compactness and in particular when it comes to develop a robotic assistant for percutaneous interventions under MRI. Indeed, MRI imposes severe design contraints in terms of compactness and compatibility. Besides, these mechanisms have the ability to adapt their stiffness level. This property is interesting to design, for instance, variable stiffness devices able to perform gestures on organs under physiological motions such as the liver without organ lacerations.

2. Thesis objective

The many aforementioned properties are of interest only if they are well exploited. This exploitation involves necessarily the control of the behaviour of these mechanisms. In the framework of this thesis, we pay a special attention to the control of tensegrity mechanisms. Indeed, very little work exists today on this topic. In the literature, two approaches to control simultaneously the position and the stiffness of a planar 1 degree of freedom tensegrity mechanism have been developed [1]. They are based on the static model inversion and the use of cable tension distribution algorithms, inspired from the control of classical cable-driven robots [2]. The results in terms of position tracking for

these approaches are satisfactory but only for slow trajectories. For high dynamic trajectories, tracking performances are very poor, limiting thus the relevance of these control approaches. Regarding the stiffness control, this was made possible thanks to the simple geometry of the mechanism. In contrast, for spatial architectures with several degrees of freedom, this requires the development of more elaborated approaches.

The objective of this thesis is therefore to develop new control approaches for tensegrity mechanisms in order to improve their performances. To this end, the PhD student has to :

  • propose a generic dynamic modelling formulation of tensegrity mechanisms ;
  • develop an identification method for parameter estimation ;
  • elaborate an advanced model-based control approach ;
  • extend the stiffness control of planar tensegrity mechanisms to spatial ones ;
  • evaluate the position/stiffness control approach on a prototype to be developed.

3. Requirements

The candidate must hold a Master’s degree (or equivalent) in October 2017 in robotics, automation or mechatronics. He must have a solid background and experience in robotics, control and real-time C/C++ programming. Experience in mechanical design is appreciated and a good command of the scientific English language is expected.

4. Contact

To apply to this offer, send an email with a CV, a cover letter and your Master grades to Salih Abdelaziz and Philippe Poignet :,


[1]  Q. Boehler S. Abdelaziz, M. Vedrines, P. Poignet and P. Renaud, From Modelin to Control of A Variable Stiffness Device Based on A Cable-driven Tensegrity Mechanism. Mechanism And Machine Theory, Vol. 102, Page 1-12, 2017

[2]  M. Gouttefarde, J. Lamaury, C. Reichert, T. Bruckmann, "A Versatile Tension Distribution Algorithm for n-DOF Parallel Robots Driven by n+2 Cables," IEEE Transactions on Robotics, Vol. 31, No. 6, pp. 1444-1457, 2015. 

Last update on 07/03/2017