Stabilizing Controllers for Tunnel Merging Maneuvers of a Multi Agent System

The ability of a robot to navigate through narrow and constrained spaces such as tunnels can be a critical requirement for many applications, including search and rescue missions, inspection tasks, and exploration of underground environments. In this paper, a multi-agent system is to navigate a constrained tunnel environment to their target locations performing obstacle and collision avoidance and tunnel merging maneuvers. Lyapunov-based control scheme (LbCS), a classical approach classified under the artificial potential fields (APFs) method, is employed to design a total potential. The total potential comprises of attractive and repulsive potential functions for target attraction, obstacle and collision avoidances, and restrictions of the individuals as artificial obstacles. A novel set of stabilizing continuous acceleration based controllers are extracted from the total potential that ensures cost and time effectiveness, safety and continuous paths for the individuals of the multi agent system. The efficacy of the controllers is verified and validated through computer simulations using Wolfram Mathematica 13.1 software. The controllers are then applied to nonholonomic car-like vehicles. Metro and subway networks can utilise this system to inspect for spalling and diagnose tunnel deterioration. The controllers derived could be easily applied to prototype robots such as miniature unmanned ground/aerial vehicles.