A Two-Stage Method for Target Searching in the Path Planning for Mobile Robots.

Area of Science:

• Robotics
• Computer Vision
• Artificial Intelligence

Background:

• Mobile robot path planning is crucial for applications like inspection and elder care.
• Achieving complete visual coverage with onboard cameras presents significant challenges.

Purpose of the Study:

• To develop an efficient two-stage optimization model for mobile robot target searching and complete visual coverage.
• To minimize search path length and search time in complex environments.

Main Methods:

• A novel method identifies key locations for visual coverage, inspired by image processing corner detection.
• A shortest path planning algorithm is designed to visit key locations, incorporating target occurrence frequency.
• The algorithm was tested against Rule-based and Genetic Algorithms (GA) in simulations and a real environment.

Main Results:

• The proposed algorithm achieved significantly shorter search path lengths and reduced target search times compared to existing methods.
• Incorporating prior target occurrence frequency further optimized search times.
• Feasibility was confirmed through real-world environment testing.

Conclusions:

• The two-stage optimization model offers an effective solution for mobile robot path planning and complete visual coverage.
• The algorithm demonstrates superior performance in efficiency and path optimization for target searching tasks.