A Cooperative Level Curve Tracking Algorithm without Communications

Algorithms for a multi-robot system to track the desired level curve of a scalar field have a wide range of applications in environmental exploration. Traditional algorithms generally require communications between robots for gradient estimation on the unknown environment. In the case where the accurate global locations and communications are limited, e.g., underwater environments, communication-based algorithms may be inapplicable. This paper proposes an improved algorithm that does not rely on communications nor global locations of robots. A distributed formation control law is designed to solve the level curve tracking problem. Each robot only uses local measurements of the field and relative positions of its neighbors to maintain formation. The dynamics and stability of the algorithm are analyzed. Additionally, the efficiency of the algorithm is validated through simulations.