Multiple trajectory optimization and control of robotic agents using hybrid fuzzy embedded artificial intelligence...

Area of Science:

• Robotics
• Artificial Intelligence
• Control Systems

Background:

• Wheeled robots are crucial for automating complex tasks, replacing human effort.
• Optimizing robotic paths and minimizing time consumption are key challenges in current automation goals.
• Path optimization and control of mobile robots remain active research areas.

Purpose of the Study:

• To develop a multi-objective path optimization technique for wheeled robots.
• To enhance navigation efficiency and reduce task completion time for mobile robots.
• To address real-time objectives in robotic path planning and control.

Main Methods:

• A hybrid approach combining fuzzy logic and a modified marine predator optimization algorithm was developed.
• Fuzzy logic processed obstacle distances, with its output feeding into the optimization algorithm.
• A Petri-Net controller was integrated to manage navigation and prevent inter-robot collisions with moving obstacles.

Main Results:

• The hybrid fuzzy-marine predator algorithm demonstrated successful navigation for multiple robots in simulations and real-time experiments.
• The proposed technique showed an average improvement of approximately 10% or more in navigational parameters compared to existing methods.
• The system effectively avoided inter-robot collisions in the presence of moving obstacles.

Conclusions:

• The hybridized fuzzy-marine predator optimization algorithm offers a robust solution for mobile robot path planning.
• The integration of a Petri-Net controller further refines navigation, especially in dynamic environments with multiple robots.
• The proposed method significantly improves navigational parameters, validating its effectiveness against other AI techniques.