Reactive multi-agent systems present global behaviours uneasily linked to their local dynamics. When it comes to controlling such a system, usual analytical tools are difficult to use so specific techniques have to be engineered. We propose an experimental dynamical approach to control the global behaviour of a reactive multi-agent system. We use reinforcement learning tools to link global information of the system to control actions. We propose to use the behaviour of the system as this global information. The controllability is evaluated in terms of rate of convergence towards a target behaviour. We compare the results obtained on a toy example with the usual approach of parameter setting.