In this paper, we investigate a novel two-echelon ground vehicle and its mounted unmanned aerial vehicle cooperative routing problem (2E-GUCRP) for intelligence, surveillance, and reconnaissance (ISR) missions. In a typical cooperative system with ground vehicle (GV) and unmanned aerial vehicle (UAV), the UAV is launched from the GV, automatically flies to the designated target and return to the GV before powering off. Meanwhile, acting as a mobile platform, the GV can carry the UAV moving in a larger region and charge or exchange the UAV's battery. The objective is to design efficient GV and UAV routes to minimize the total mission time while meeting the operational constraints. A mixed integer programming (MIP) model, which could be solved by commercial software (e.g. CPLEX), is developed to formulate the problem. In order to quickly solve the medium-scale or above problems, two heuristics are designed. The random instances and practical case based on the road network in Changsha, China, are used for experiments. The computational results indicate that, compared to the mode of two-echelon vehicle routing problem (2E-VRP), the cooperative GV and UAV can complete the ISR missions in a much shorter time through utilizing the proposed 2E-GUCRP mode. Furthermore, discussion of the practical instance points out that the increase in efficiency is related to the speed relationship between the GV and the UAV. INDEX TERMS Two-echelon routing, vehicle routing, UAV routing, heuristic.