Homepage of Mazen Alamir: Works in Robotics and Mechatronics

Fast MPC Swing-up and Stabilization of a Twin Pendulum Systems

Joint work with A. Murilo and G. Buche (Gipsa-lab)

Further Results on the Accessibility of a Satellite with Two Reaction Wheels

This article deals with the attitude open loop control of a satellite actuated by two reaction wheels and contributes to answer to the question of [1]: “what is the most one can expect in such a case?” As a matter of fact, the satellite with two reaction wheels is not controllable on the whole state space of the spacecraft attitudes. Nevertheless, a five dimensional sub-manifold is susceptible to be accessible from any state. In this article, this sub-manifold will be exhibited and two control objectives related to it, will be studied. The first one corresponds to steer the satellite from any attitude state to any state at rest. The second one corresponds to steer the spacecraft from any state to a given configuration with a constant spin about the unactuated spacecraft axis. More than giving qualitative answers, the article gives constructive demonstrations of accessibility since it exhibits explicit open loop control laws based on a path planning compatible with these two objectives. Finally, some simulation results illustrate the feasibility of the approach.

Reference

Boyer, F.; Alamir, M. Further results on the accessibility of a satellite with two reaction wheels. Journal of Guidance, Control, and Dynamics. Vol 30(2) (2007). [download]

Multi-step limit cycle generation for Rabbit’s walking based on a nonlinear low dimensional predictive control scheme

In this paper, a new nonlinear predictive control scheme is proposed for a five-link planar under-actuated biped walking robot. The basic feature in the proposed strategy is to use on-line optimization to update the tracked trajectories in the completely controlled variables (actuated coordinates) in order to enhance the behavior and the stability of the remaining indirectly controlled ones (unactuated coordinates). The stability issue is discussed using the Poincare ́’s section tool leading to a computable criterion that enables the stability of the overall scheme to be investigated as well as the computation of a candidate region of attraction. The whole framework is illustrated through simulation case-studies. To attest the efficiency of the proposed scheme, robustness against model uncertainties and ground irregularities are investigated by simulation studies

Reference

[24] Chemori, A.; Alamir, M. Multi-step limit cycle generation for Rabbit's walking based on a nonlinear low dimensional predictive control scheme. International Journal of Mechatronics, 16, No.5, 259-277 (2006). [download]

Fast Generation of Attractive Trajectories for an Under-Actuated Satellite Application to Feedback Control Design

An algorithm is proposed that perfonns a fast generation of trajectories that steer the state of an under- actuated satellite from a given initial value to the neighborhood of a final desired one. The proposed algorithm is then used to design a sampled-data state feedback control. Finally, simulations are proposed to test the robustness of the closed-loop behaviour under several classes of uncertainties and disturbances.

Reference

Alamir, M.; Boyer, F. Fast generation of attractive trajectories for an under-actuated satellite application to feedback control design. Optim. Eng. 4, No.3, 215-230 (2003). [download]

Constrained Minimum-Time-Oriented Feedback Control for the Stabilization of Nonholonomic Systems in Chained Form

In this paper, a discontinuous state-feedback law is pro- posed for the stabilization of nonholonomic systems in power form. The feedback law is based on a receding-horizon strategy in which the open-loop optimization problem is a minimum-time steering process. Suboptimal formulations are used explicitly to meet the real-time implementability requirements. Stability is established in a sampled- data context and illustrative simulations are given to show the effective- ness and the real-time implementability of the proposed scheme.

Reference

Alamir, M.; Marchand, N. Constrained minimum-time-oriented feedback control for the stabilization of nonholonomic systems in chained form. J. Optimization Theory Appl. 118, No.2, 229-244 (2003). [download]

Robot Anguille sous marin en 3D

Reference

Boyer, F.; Alamir, M.; Chablat, D.; Khalil, W.; Leroyer, A.; Lemoine, Ph. Robot Anguille sous marin en 3D. "Techniques de l'ingénieur", S7856, pp 1-16, (2006).

Re-injecting the structure in NMPC schemes: Application to the constrained stabilization of a snakeboard

In this paper, a constrained nonlinear predictive control scheme is pro- posed for a class of under-actuated nonholonomic systems. The scheme is based on fast generation of steering trajectories that inherently fulfill the contraints while showing a ”translatability” property which is generally needed to derive stability re- sults in receding-horizon schemes. The corresponding open-loop optimization prob- lem can be solved very efficiently making possible a real-time implementation on fast systems (The resulting optimization problem is roughly scalar). The whole frame- work is shown to hold for the well known challenging problem of a snakeboard con- strained stabilization. Illustrative simulations are proposed to assess the efficiency of the proposed solution under saturation constraints and model uncertainties.

Reference

Alamir, M.; Boyer, F. Re-injecting the Structure in NMPC Schemes - Application to the Constrained Stabilization of a Snakeboard. In Fast Motions in BioMechanics and Robotics, Optimization and Feedback Control. Lecture Notes in Control and Information Sciences, Springer-Verlag, (2006). [download]

On Solving Inverse Problems for Electric Fish Like Robots

This paper relates preliminary results concerning the solution of inverse problems arising in electric sense based navigation. This sense is used by electric fishes to move in dark waters using the electric current measurements perceived by the epidermal sensors as these are affected by the presence of obstacles. The latter change the resulting induced measures by instantaneously disturbing the fish self-produced electric field. The approach lies on a recently proposed graphical signature based classification methodology to overcome the computational burden associated to an explicit inversion of the mathematical equations. A preliminary validation of the proposed solution is obtained using a dedicated experimental setting.

Reference

M. Alamir, O. Omar, N. Servagent, A. Girin, P. Bellemain, V. Lebastard, P. B. Gossiaux, F. Boyer and S. Bouvier. On Solving Inverse Problems for Electric Fish Like Robots. Proceedings of the 2010 IEEE Conference on Robotics and Biomimetics, Tianjin, China, 2010. [download]

Reduced Mean Model for Controlling a Three-Dimensional Eel-like Robot

This paper presents a reduced mean model of a three-dimensional Eel-like robot. Such a robot is under construction in the context of a national French robotic project. This model is based on mechanical considerations as well as on our experience with an existing 3D continuous model of the target prototype. Identification and validation of the dynamic model are presented here.

Reference

El-Rafei M., M. Alamir, N. Marchand, M. Porez and F. Boyer. Reduced Mean Model For Controlling a Three-Dimensional Eel-Like Robot. Proceedings of the IEEE Conference on Robotics and Biomimetics, Bangkok, Thailand 2008. [download]

Feedback design for 3D movement of an Eel-like robot

This paper relates recent advances in the design of feedback laws for the 3D movement of an Eel-like robot. Such a robot is under construction in the context of a national French robotic project. The proposed feedback enables the tracking of a desired 3D position of the Eel head as well as the stabilization of the rolling angle. A velocity controller is also proposed. The controller is tested on a recently developed complete 3D model in order to assess its efficiency in tackling 3D manoeuvres.

Reference

Alamir M., El Rafei M., Hafidi G., Marchand N., Porez M. and Boyer F. Feedback Design of a 3D Movement of an Eel Robot. Proceedings of the IEEE International Conference on Robotics and Automation, ICRA'07 (2007). [download]

Motion Control of a Three-Dimensional Eel-like Robot Without Pectoral Fins

In this paper, recent advances in the design of feedback laws for the 3D movement of an Eel-like robot are presented. Such a robot is under construction in the context of a national French robotic project. The proposed feedback enables the tracking of a desired 3D position of the Eel head as well as the stabilization of the rolling angle without using pectoral fins. We build on a previous work in which we proposed a complete control scheme for robot’s 3D movement using its pectoral fins. The controller is tested on a recently developed complete 3D model in order to assess its efficiency in tackling 3D manoeuvres

Reference

El Rafei M., Alamir M., Marchand N., Porez M. and Boyer F. Motion control of a three dimensional eel-like robot without pectorial fins. Proceedings of the IFAC World Congress, South Korea (2008). [download]

Multi-variable Constrained Control Approach for a Three-Dimensional Eel-like Robot

In this paper, a multi-variable feedback design for the 3D movement of an eel-like robot is presented. Such a robot is under construction in the context of a national French robotic project. The proposed feedback enables the tracking of a desired 3D position of the eel head as well as the stabilization of the rolling angle. The control design is based on a recently developed reduced model that have been validated using a 3D complete continuous model described in [3]. Several scenarios are proposed to assess the efficiency of the proposed feedback law.

Reference

El Rafei M., Alamir M., Marchand N., Porez M. and Boyer F. Multi-variable Constrained Control Approach for a Three-Dimensional Eel-Like Robot. Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, (2008). [download]

On friction compensation without friction model

This paper presents a new methodology for friction compensation that is not based on any friction model. This is done by using finite-terms Fourier series to approx- imate the friction term. Updating laws for the coefficients of the series are eas- ily derived from a Lyapunov approach to guarantee asymptotic convergence of the tracking error. Computer simulations are then proposed in which friction term is generated by existing models (unknown to the controller) to illustrate the efficiency of the proposed approach in compensating friction effects without friction model. Extension to the case of unknown inertia is also proposed. Robustness against ve- locity measurement errors is investigated by simulation. Moreover, since no friction model is used in the proposed compensation scheme, the resulting controller can be used without any preliminary identification phase necessary in most of the existing approaches.

Reference

Alamir, M. On friction compensation without friction model. Proceedings of the IFAC World Congress, Barcelona, Spain (2002). [download]

A New Control Architecture For Multi-Beam Fringe Tracker

Fringe-tracking has long been recognized as a critical system for modern astronomical ground interferometry to stabilise observations. The incoming generation of trackers is intended to co-phase a large number of telescopes simultaneously, bringing new questions related to control and redundancy. In this paper, we propose a new control architecture for the 4/6-telescope Second Generation Fringe Tracker, currently under study for the Very Large Telespoes Interferometer (VLTI). The main feature of the proposed solution lies in the explicit handling of coupling and redundancy. This enables in particular to tune the different baselines control-related weighting coefficients according to the presence of noise and or potential loss of flux. Moreover, the unavoidable delays are explicitly taken into account in the control design while an observer is used to reconstruct the dynamic of the atmospheric OPD. The control design is based on a multi-variable state space representation making possible the use of standard Linear Quadratic design. The resulting controller can still be expressed in a transfer function form. The efficiency of the proposed solution is illustrated through dedicated simulations involving realistic data.

Reference

L. Vincent, M. Alamir, J. B. Le Bouquin, K. Perraut, P. Kern, L. Jocou and J. P. Berger. A New Control Architecture for Multi-Beam Fringe Tracker. SPIE Symposium on Astronomical Telescopes and Instrumentation: Optical and Infrared Interferometry II. San Diego, USA, June 2010. [download]