A Hybrid Path-Planning Strategy for Mobile Robots with Limited Sensor Capabilities.

Area of Science:

• Robotics
• Artificial Intelligence
• Computer Science

Background:

• Path planning is crucial for autonomous systems.
• Limited sensor and processing capabilities pose challenges for mobile robots.
• Existing algorithms often require prior environmental knowledge or significant computational resources.

Purpose of the Study:

• To develop a novel, resource-efficient path planning strategy for platforms with constrained capabilities.
• To create a hybrid approach integrating global and local planning methods.
• To validate the algorithm's effectiveness in real-world, cluttered environments.

Main Methods:

• A hybrid path planning algorithm combining a global planner for known map areas and a local planner for real-time obstacle avoidance.
• The algorithm operates without prior environmental information, relying on an integrated mapping capability.
• Utilized a Pioneer P3-DX robot equipped with an Xbox 360 Kinect sensor for experimental validation.

Main Results:

• The proposed hybrid algorithm successfully generated sub-optimal, collision-free paths in a known map.
• Real-time obstacle avoidance was achieved efficiently using the cost-effective local planner.
• Experimental validation demonstrated the algorithm's practical applicability and efficiency in a cluttered environment.

Conclusions:

• The developed hybrid path planning strategy is effective for resource-constrained platforms.
• The integration of global and local planning offers a robust solution for autonomous navigation.
• The algorithm's efficiency and adaptability make it suitable for real-world robotic applications.