Hybrid Swarm Intelligence and Human-Inspired Optimization for Urban Drone Path Planning

Area of Science:

• Robotics and Automation
• Artificial Intelligence
• Computational Optimization

Background:

• Urban drone operations necessitate advanced path planning for safe and efficient navigation.
• Existing algorithms often struggle with complex urban environments, leading to suboptimal routes and potential safety concerns.

Purpose of the Study:

• To develop an enhanced path planning algorithm for urban drones.
• To improve navigation safety, efficiency, and trajectory smoothness in complex urban settings.

Main Methods:

• Integration of hierarchical structures and group interaction behaviors into Particle Swarm Optimization (PSO).
• Mathematical modeling of competitive and supportive behaviors to enhance PSO's learning and global search.
• Introduction of a mutation mechanism to prevent local optima convergence and improve diversity.
• Combination of path segmentation, prioritized update, and cubic B-spline curve algorithms for path generation.

Main Results:

• The proposed method generates smoother drone trajectories compared to standard approaches.
• Demonstrated improvements in real-time performance and significant reductions in energy consumption and operation time.
• Effectiveness validated through comparative simulations in complex urban environments.

Conclusions:

• The novel approach significantly advances urban drone path planning capabilities.
• The method ensures safe, optimal, and efficient drone navigation, broadening applicability in diverse urban scenarios.