Bézier Curves-Based Optimal Trajectory Design for Multirotor UAVs with Any-Angle Pathfinding Algorithms.

Area of Science:

• Robotics
• Artificial Intelligence
• Aerospace Engineering

Background:

• Multirotor Unmanned Aerial Vehicles (UAVs) are crucial for applications in complex, obstacle-rich environments.
• Planning safe and efficient flight paths for UAVs in cluttered spaces presents significant challenges.
• Existing pathfinding algorithms require adaptation to account for environmental complexities and mission constraints.

Purpose of the Study:

• To develop and evaluate methods for generating flyable and continuous trajectories for multirotor UAVs in cluttered environments.
• To modify and apply any-angle pathfinding algorithms (Basic Theta*, Lazy Theta*, Phi*) with an elevation-aware cost function.
• To propose and assess Bézier curve-based approaches for smoothing paths into optimal, collision-free trajectories.

Main Methods:

• Modified Basic Theta*, Lazy Theta*, and Phi* algorithms incorporating an elevation-aware cost function to generate initial paths.
• Four Bézier curve-based methods applied to smooth the generated paths into continuous, flyable trajectories.
• An innovative evaluation process to assess the success rates of different smoothing approaches in diverse environments.

Main Results:

• The modified pathfinding algorithms successfully generated suitable paths in cluttered environments.
• Bézier curve-based smoothing approaches demonstrated high success rates in creating flyable trajectories.
• An evaluation process identified the most effective smoothing approach for optimal trajectory design.

Conclusions:

• The integration of modified any-angle pathfinding algorithms and Bézier curve smoothing offers an effective solution for UAV trajectory planning.
• The proposed evaluation framework is valuable for selecting optimal trajectory generation methods.
• A novel algorithm was developed to enhance the efficiency and optimality of the selected trajectory construction approach.