Planning for Autonomous Unmanned Ground Vehicles (AUGV) is still a challenge,\nespecially in difficult, off-road, critical situations. Automatic planning can\nbe used to reach mission objectives, to perform navigation or maneuvers. Most\nof the time, the problem consists in finding a path from a source to a\ndestination, while satisfying some operational constraints. In a graph without\nnegative cycles, the computation of the single-pair shortest path from a start\nnode to an end node is solved in polynomial time. Additional constraints on the\nsolution path can however make the problem harder to solve. This becomes the\ncase when we need the path to pass through a few mandatory nodes without\nrequiring a specific order of visit. The complexity grows exponentially with\nthe number of mandatory nodes to visit. In this paper, we focus on shortest\npath search with mandatory nodes on a given connected graph. We propose a\nhybrid model that combines a constraint-based solver and a graph convolutional\nneural network to improve search performance. Promising results are obtained on\nrealistic scenarios.\n