Dynamic Coordination: Advancing Multi-Robot Vehicle Control Through Nonlinear Model Predictive Controllers

Abstract

Dynamic coordination in multi-robots aims to achieve autonomous navigation with an effective obstacle avoidance mechanism. One of the main challenges associated with multi-robot coordination is handling the changes in the unknown environments. It is important to design an efficient coordination system which can provide reliable and safe path planning for a multi-robot system in a dynamic environment. This research presents an advanced approach for multi-robot coordination using nonlinear model predictive control (NLMPC) framework. Unlike linear MPCs, NLMPCs are highly effective in controlling nonlinear dynamics such as input/output constraints and robot parameters. In this research, the NLMPC strategy is used to train the robots to find targets and reach the destination by avoiding obstacles in the dynamic environment. The coordination mechanism and path planning approach employed in this research enables the robots to successfully search for the target and navigate through the obstacles. The proposed coordination approach follows a specific trajectory and adjusts the vehicle control parameters such as acceleration, and steering angle in order to stay on the specified path. Simulation is conducted to visualize the motion of the multi-robot vehicle through different trajectories. Results of the simulation show that the proposed strategy exhibits excellent acceleration and steering control and successfully avoids obstacles.