Behaviour-defined navigation framework for dynamical obstacle avoidance in multi-robot systems consisting of holonomic robots

Dynamical obstacle avoidance is a challenging problem in the field of autonomous robot navigation. Current research in this field has been mostly limited to single robots, thus, there exists a gap in research in the field of dynamical obstacle avoidance in multi robot systems. While rich literature is available on multirobot systems, this paper attempts to propose a novel navigation framework for an environment which includes multiple robots. The proposed navigation framework applies certain behaviours to ensure a safe trajectory for the multi-robot systems. As opposed to other reported literature which focused on implementing their algorithms on non-holonomic robots, the proposed navigation framework is implemented on several holonomic robots. Simulations and real-life experiments were carried out using the proposed framework. Dynamic obstacles are considered in the environment and Khepera IV robots are used to conduct real-life experiments. Two dynamic obstacles were placed at different positions in the workspace. These obstacles had linear movements, whereby each robot could move horizontally and vertically across the workspace. Three experimental trials were performed. Results show that the proposed navigation framework is successful in navigating the multi-robot system to their respective target locations while avoiding obstacles.