Microrobot Path Planning Based on the Multi-Module DWA Method in Crossing Dense Obstacle Scenario.

Area of Science:

• Microrobotics
• Path Planning
• Obstacle Avoidance

Background:

• The Dynamic Window Approach (DWA) is a common path planning algorithm but struggles with complex, dense obstacle environments.
• Microrobot navigation in cluttered spaces presents significant challenges for existing algorithms, leading to low success rates.

Purpose of the Study:

• To develop a novel multi-module enhanced Dynamic Window Approach (MEDWA) for improved microrobot path planning.
• To enhance microrobot success rates and efficiency in navigating complex, obstacle-dense environments.

Main Methods:

• A multi-obstacle coverage model incorporating Mahalanobis distance, Frobenius norm, and covariance matrix identifies obstacle-dense areas.
• MEDWA combines enhanced DWA (EDWA) for non-dense areas and 2D analytic vector field methods for dense areas.
• EDWA utilizes a modified navigation function and dynamically adjusted weights via an improved immune algorithm (IIA) for trajectory optimization.

Main Results:

• MEDWA demonstrated a significant improvement in microrobot ability to traverse obstacle-dense areas.
• The proposed method reduced trajectory length and the number of steps by approximately 15%.
• Planning deviation was minimized, and microrobots successfully avoided obstacles without deviation or collision.

Conclusions:

• MEDWA effectively addresses the limitations of traditional DWA in complex, dense obstacle scenarios.
• The hybrid approach enhances microrobot navigation capabilities, offering a more robust and efficient path planning solution.
• The algorithm's adaptability and optimization contribute to successful microrobot operations in challenging environments.