This paper deals with problems of sensor management in a human driven information collection process. This applicative context results in complex sensor-to-task assignment problems, which encompass several difficulties. First of all, the tasks take the form of several information requirements, which are linked together by logical connections and priority rankings. Second, the assignment problem is correlated by many constraint paradigms. Our problem is a variant of Vehicle Routing Problem with Time Windows (VRPTW), and it also implements resource constraints including refuelling issues. For solving this problem, we propose a column generation approach, where the label correcting method is used to treat the sub-problem. The efficiency of our approach is evaluated by comparing with solution given by CPLEX on different scenarios.