Source code for "Designing Routes for a Fleet of Drones under Dynamic Conditions: Combining heuristics with reinforcement learning"

This code addresses the Dynamic Team Orienteering Problem. This problem considers a limited fleet of drones, and aims to design efficient routes, maximizing the total reward obtained from visiting a set of nodes while considering battery constraints and dynamic rewards that depend upon the contextual conditions. To solve this problem, a learnheuristic algorithm combining a constructive heuristic with a reinforcement learning method is presented.