PROJECT TITLE :
Automation-Assisted Capture-the-Flag: A Differential Game Approach
Capture-the-flag could be a complicated, difficult game that is a helpful proxy for several issues in robotics and other application areas. The game is adversarial, with multiple, probably competing, objectives. This interplay among different factors makes the matter complex, even within the case of solely 2 players. To form analysis tractable, previous approaches typically create various limiting assumptions upon player actions. In this paper, we tend to gift a framework for analyzing and solving a 2-player capture-the-flag game as a zero-total differential game. Our drawback formulation allows each player to make decisions rationally in keeping with this player positions, assuming only an higher sure on the movement speeds. Using numerical solutions to Hamilton-Jacobi-Isaacs equations, we have a tendency to compute winning regions for every player as subsets of the joint configuration space and derive the corresponding winning ways. The computational method and simulations are presented, together with experiments with human agents within the Berkeley autonomy and robotics in capture-the-flag testbed. These experiments demonstrate the employment of the solutions in realistic conditions and highlight their potential applications in automation-aided decision creating for humans and mixed human-robot groups.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here