Fault Point Search with Obstacle Avoidance for Machinery Diagnostic Robots Using Hierarchical Fuzzy Logic Control

Area of Science:

• Robotics and Automation
• Artificial Intelligence
• Industrial Engineering

Background:

• Modern factories require advanced fault detection for continuous operation.
• Manual inspection is time-consuming, necessitating autonomous diagnostic robots.
• Existing robots face safety limitations in dynamic factory environments.

Purpose of the Study:

• To propose a hierarchical fuzzy logic-based navigation and obstacle avoidance algorithm for diagnostic robots.
• To enhance the safety and efficiency of autonomous fault detection in factories.
• To address the limitations of current diagnostic robot navigation systems.

Main Methods:

• Developed a hierarchical fuzzy logic algorithm using zero-order Takagi-Sugeno fuzzy control.
• Integrated subfunctions for navigation, static, and dynamic obstacle avoidance.
• Introduced a dynamic safety boundary concept using normalized breached level and relative speed direction for obstacle avoidance.

Main Results:

• The proposed algorithm successfully guided the diagnostic robot to within 30 cm of the fault point.
• Ensured collision avoidance with both static equipment and dynamic obstacles.
• Demonstrated enhanced completeness and safety in the fault point searching process through simulations.

Conclusions:

• The hierarchical fuzzy logic algorithm provides a safe and effective solution for autonomous diagnostic robot navigation.
• The dynamic safety boundary and multi-objective planning enhance robot performance in complex industrial settings.
• This approach improves the overall safety and efficiency of fault detection in continuously operating machines.