Indoor Visual Exploration with Multi-Rotor Aerial Robotic Vehicles.

Area of Science:

• Robotics
• Artificial Intelligence
• Control Systems

Background:

• Autonomous exploration of unknown indoor environments presents significant challenges for multi-rotor robots.
• Existing methods often struggle with safety guarantees, computational complexity, and multi-agent coordination.

Purpose of the Study:

• To develop a reactive algorithm for autonomous exploration using multiple autonomous multi-rotor robots in indoor, unknown environments.
• To ensure provable safety and complete workspace exploration through a novel control architecture.

Main Methods:

• A two-level control architecture combining Artificial-Harmonic Potential Field (AHPF) for navigation and a low-level tracking controller.
• Extension to a multi-agent approach for collaborative navigation.
• Integration of approximate techniques for AHPF acquisition to reduce computational complexity.

Main Results:

• The AHPF guarantees safety and workspace exploration.
• The low-level controller ensures safe tracking of the AHPF via velocity commands.
• High-fidelity simulations validated the robust framework for multi-agent exploration tasks.

Conclusions:

• The proposed method offers a robust and computationally efficient solution for multi-agent autonomous exploration in unknown indoor environments.
• The two-level control architecture effectively addresses the challenges of navigation, safety, and coordination for multi-rotor robots.