CAMO: A Conditional Neural Solver for the Multi-objective Multiple Traveling Salesman Problem

This paper presents CAMO, an AI system that helps teams of robots efficiently visit multiple locations while balancing competing goals like minimizing travel time and total distance. The system uses a neural network with a special design that can handle different numbers of robots and targets, and it learns to find multiple good solutions that represent different trade-offs between the competing objectives. The researchers tested it on both simulated problems and real robots, showing it outperforms existing methods.