Résumé
This paper presents an optimal formulation for sequencing a set of robotic tasks. Tasks description and execution are based on the task formalism. A naive solution would be to arrange the execution of the tasks sequentially (potential redundancy not exploited, not optimal), or to set manually the timing and the behavior for each task (burdensome for the user, lack of autonomy, not optimal), or to arrange the tasks by priority and pill them up at once in the stack-of-tasks (high likelihood of conflicts, wrong or cumbersome execution). Our contribution is to determine the time-optimal realization of the mission taking into account robotic constraints as complex as collision avoidance; its originality lies in keeping the task formalism in the formulated optimization of the task sequencing problem. This theory is exemplified through a simulation on a 2D robot.