Monte Carlo Tree Search Based Tactical Maneuvering
This work addresses tactical maneuvering for unmanned aircraft, but it is incremental as it applies an existing MCTS method to a new domain without major innovations.
The paper tackles tactical maneuvering between two unmanned aircraft by applying a simultaneous move Monte Carlo Tree Search (MCTS) framework, demonstrating its effectiveness in a simulated 2D application with efficient long-horizon search and self-play for maneuver selection.
In this paper we explore the application of simultaneous move Monte Carlo Tree Search (MCTS) based online framework for tactical maneuvering between two unmanned aircrafts. Compared to other techniques, MCTS enables efficient search over long horizons and uses self-play to select best maneuver in the current state while accounting for the opponent aircraft tactics. We explore different algorithmic choices in MCTS and demonstrate the framework numerically in a simulated 2D tactical maneuvering application.