Efficient Informative Path Planning via Normalized Utility in Unknown Environments Exploration.

Area of Science:

• Robotics
• Artificial Intelligence
• Computer Science

Background:

• Autonomous exploration is crucial for unmanned aerial vehicles (UAVs) in tasks like search and rescue.
• Efficiently mapping unknown environments remains a significant challenge in robotics.

Purpose of the Study:

• To develop an efficient exploration framework for UAVs in unknown 3D environments.
• To create a method for planning informative paths that balance data acquisition and travel cost.

Main Methods:

• Incrementally building and maintaining a topological roadmap using Rapidly-exploring Random Trees (RRT).
• Utilizing a normalized utility function to evaluate path informativeness based on map uncertainty and distance cost.
• Employing a local planner to determine optimal paths for data collection and map updating.

Main Results:

• The proposed framework demonstrated superior exploration efficiency compared to existing methods in simulated 3D environments.
• The topological roadmap effectively guided the UAV to search for optimal informative paths.
• The system successfully collected measurements and restructured the map during exploration.

Conclusions:

• The developed framework offers an efficient solution for autonomous exploration using UAVs.
• The integration of topological mapping and informative path planning enhances exploration performance.
• This approach is valuable for reconnaissance and search missions in unknown terrains.