Priority Task-Based Formation Control and Obstacle Avoidance of Holonomic Agents with Continuous Control Inputs

摘要

This work addresses the problem of controlling a network of connected and holonomic agents to achieve a formation while obstacles are avoided. The agents themselves can be obstacles for each other or there may be fixed obstacles in the environment. The proposed control law is computed by the adaptive convex combination of two control laws dealing with two tasks, one devoted to achieve a desired formation and the other focused on obstacle avoidance. The adaptive convex combination is twofold, in one hand it prioritizes the obstacle avoidance task when an agent is in the neighborhood of an obstacle, and in the other hand it maintains a continuous control law. Moreover, the formation is formulated as a consensus problem and a novel finite time control law is used to solve this problem.