[Multi-objective optimization for fuzzy controller in neuro-muscular rehabilitation]

The multi-objective optimization problem could be generally formulated as minimization of vector objectives subject to a number of constraints and bounds. Among the non-derivative problems, genetic algorithms have been proved to be a good solution in addressing ill-posed problem. We applied the multi-objective optimization to fuzzy controller design in order to obtain a optimal solution that is often a compromise between different objectives. The application that are targeted in this paper is the e-health in neuromuscular rehabilitation using electromyographies. Because our application is implemented using a micro-controller with limited capabilities, that is flash memory, computational effort, and the clock frequency, the interest is to design the smallest fuzzy architecture possible in order to accomplish the objectives. The optimal set was used in order to achieves these objectives.