Artificial Intelligence Algorithms in Behavioural Control of Wheeled Mobile Robots Formation

摘要

The paper presents an innovative approach to the problem of the wheeled mobile robots formation behavioural control with use of artificial intelligence algorithms. The control task is solved by application of adaptive dynamic programming algorithms in the hierarchical control system, that generates the collision free trajectories in the unknown 2D environment for all agents in the formation, and realises generated trajectories using tracking control algorithms. The hierarchical control system consists of three layers: the trajectory generator, the wheeled mobile robots formation control system and tracking control systems for individual agents. The trajectory generator presents the new approach to the behavioural control, where one neural dynamic programming algorithm generates the behavioural control signals that make possible to compute the trajectory for realisation of the complex task, which is a composition of two individual behaviours: "goal-seeking"and "obstacle avoiding". Computer simulations have been conducted to illustrate the path planning process.