An Interpretable Fuzzy System Learned Through Online Rule Generation and Multiobjective ACO With a Mobile Robot...

Area of Science:

• Robotics
• Artificial Intelligence
• Control Systems Engineering

Background:

• Designing interpretable fuzzy logic systems (FLS) with high control performance is challenging.
• Existing methods often require offline training data, which is not always available.
• Online generation of FLS rules and parameters is crucial for adaptive systems.

Purpose of the Study:

• To propose a novel multiobjective optimization approach for designing FLS using online process data.
• To apply this approach to the wall-following control of a mobile robot.
• To simultaneously optimize FLS interpretability and control performance.

Main Methods:

• An online clustering and fuzzy set merging (OCFM) algorithm generates a reference rule base.
• A multiobjective front-guided continuous ant-colony optimization (MO-FCACO) algorithm optimizes FLS structure and parameters.
• Transparency-oriented and control performance objective functions are defined and optimized.

Main Results:

• The MO-FCACO algorithm effectively optimizes FLS parameters for wall-following tasks.
• The designed FLS demonstrated effective control of a real mobile robot's orientation and speed.
• Optimization performance was validated against other multiobjective algorithms.

Conclusions:

• The proposed multiobjective FLS design approach is effective for mobile robot control.
• The method successfully balances interpretability and performance without offline data.
• The MO-FCACO algorithm provides a robust optimization framework for FLS design.