Research on autonomous navigation of mobile robots based on IA-DWA algorithm

Area of Science:

• Robotics
• Artificial Intelligence
• Navigation Systems

Background:

• Mobile robot navigation requires efficient and accurate path planning to avoid obstacles.
• Existing algorithms like A* and Dynamic Window Approach (DWA) have limitations in speed and global optimality.
• Accurate robot positioning is crucial for reliable path execution.

Purpose of the Study:

• To develop an integrated algorithm (IA-DWA) combining A* and DWA for improved mobile robot path planning.
• To enhance the speed, efficiency, and collision-free path generation capabilities.
• To improve robot positioning accuracy using sensor fusion.

Main Methods:

• Fused odometer and inertial measurement unit (IMU) data with an extended Kalman filter (EKF) for precise robot positioning.
• Optimized A* algorithm parameters (prediction function, weights, neighborhood, smoothing) to integrate global path information into DWA.
• Simulated and experimentally validated the IA-DWA algorithm on a Robot Operating System (ROS) mobile robot platform.

Main Results:

• The IA-DWA algorithm reduced path planning time by 23.3% compared to A*-DWA.
• Path length was reduced by 1.8% with IA-DWA, demonstrating improved efficiency.
• Faster convergence of optimization iterations was observed.
• Experimental validation confirmed the algorithm's reliability in autonomous navigation.

Conclusions:

• The proposed IA-DWA algorithm significantly improves mobile robot path planning efficiency and accuracy.
• Sensor fusion (EKF) enhances robot positioning, crucial for real-world navigation.
• The integrated approach effectively balances global path optimality with local obstacle avoidance.