Nacim Ramdani

SWIM 2010, the 3rd edition of the Small Workshop on Interval Methods will be held in the University of Nantes (France) on June 15th and 16th, 2010. The goal of the SWIM 2010 workshop is to bring together researchers and practitioners working on interval methods and their applications, in the broader sense, providing a forum to review and discuss the state-of-the-art in this area and fostering cross-fertilization between different approaches.

The SWIM 2010 workshop is intended to build on the success of previous editions, held in Montpellier in 2008 and Lausanne in 2009, which was initiated by the french MEA working group on Set Computation Techniques of the French research group on Automatic Control GDR MACS. The MEA group aims at promoting interval analysis, constraint propagation and other set methods to apply them in estimation, robotics, control, planning, design, verification, computer aided proof, and other engineering, computer science and applied mathematics areas.

Visit web site and submit a talk

Using hybrid automata for set-membership state estimation with uncertain nonlinear continuous-time systems, by N.Meslem, N.Ramdani & Y.Candau, published in Journal of Process Control 20(4): 481-489, 2010.

The original publication is available at www.sciencedirect.com, here doi: 10.1016/j.jprocont.2010.02.001.

V. Bonnet, P. Fraisse, N.Ramdani, J.Lagarde, S.Ramdani, B.G.Bardy (2009) A closed loop musculoskeletal model of postural coordination dynamics, 48th IEEE Conference on Decision and Control, Shanghai, China,  pp.6207-6212.

Abstract: A closed-loop model with actuator dynamics and sensory feedback has been developed to capture the complex postural behaviors observed in a human head tracking task. In motor-control litterature, spindle feedback gains are scaled by the central nervous system to adapt muscle stiffness depending on the postural task. We propose to identify spindle reflex equivalent feedback gains for several target’s frequency values. Comparison with experimental results on human shows the relevance of this modeling, since our musculoskeletal model is able to exhibit reasonably well the behavioral invariants observed in human postures.

Finalist of RoboCup Best Paper Award. Our paper S.Lengagne, N.Ramdani, P.Fraisse (2009) Planning and Fast Re-Planning of Safe Motions for Humanoid Robots : Application to a Kicking Motion, in 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, St Louis, MO, USA. pp. 441–446.

Abstract: Optimal motions are usually used as joint reference trajectories for repetitive or complex motions. In the case of soccer robots, the kicking motion is usually a benchmark motion, computed off-line, without taking into account the current position of the robot or the direction of the goal. Moreover, robots must react quickly to any situation, even if not expected, and cannot spend time to generate a new optimal motion by the classical way. Therefore, we propose a new method for fast motion re-planning based on an off-line computation of a feasible sub-set of the motion parameters, using Interval Analysis.

See related videos:
Replanned kicking motion : HOAP-3 humanoid robot kicks a ball. Database kick motion (left) was planned for ball location x=1cm, kick height h=1cm. Actual ball is located x=3cm, the ball goes slower (center). Kick motion is replanned in 1.5s CPU time (right), the ball goes faster.
Global navigation using databased motions with the HOAP-3 humanoid robot The HOAP-3 robots follows a moving target, while using safe and balanced motions, databased prior to experiment.

Visit sourceforge page for more videos and download.

Abstract: We address nonlinear reachability computation for uncertain monotone systems, those for which flows preserve a suitable partial orderings on initial conditions. In a previous work [Ramdani, 2008], we introduced a nonlinear hybridization approach to nonlinear continuous reachability computation. By analysing the signs of off-diagonal elements of system’s Jacobian matrix, a hybrid automaton can be obtained, which yields component-wise bounds for the reachable sets. One shortcoming of the method is induced by the need to use whole sets for addressing mode switching. In this paper, we improve this method and show that for the broad class of monotone dynamical systems, component-wise bounds can be obtained for the reachable set in a separate manner. As a consequence, mode switching no longer needs to use whole solution sets. We give examples which show the potentials of the new approach.

The original publication is available at www.sciencedirect.com, here doi:10.1016/j.nahs.2009.10.002.

A hybrid bounding method for computing an over-approximation for the reachable space of uncertain nonlinear systems, by N.Ramdani, N.Meslem & Y.Candau, published in IEEE Tran. Automatic Control 54(10): 2352–2364, 2009.

Abstract: This paper proposes a hybrid bounding method to compute reachable sets of continuous-time dynamical systems with parameter uncertainty. First, two appro aches to the computation of bounding boxes for state trajectories are presented for systems which are monotonic with respect to states and parameters. For general systems, a hybridization technique is used for splitting the state space into regions where the system is monotone with respect to states and parameters. The bounding systems are modeled as hybrid automata. A rule is proposed for building these automata and this paper shows howto compute an over-approximation for the reachable set. A stability analysis is given for the enclosures’ size computed by the proposed hybrid bounding method. Discussion on computational complexity and two examples of biological processes illustrate the usefulness of the proposed approach.

The original publication is available at ieeexplore.ieee.org, here doi:10.1109/TAC.2009.2028974.