Online learning fuzzy echo state network with applications on redundant manipulators

Area of Science:

• Robotics and Control Systems
• Artificial Intelligence
• Machine Learning

Background:

• Redundant manipulators are crucial for efficiency but pose challenges in inverse kinematics.
• Controlling redundant manipulators requires sophisticated methods to handle their complex joint configurations.

Purpose of the Study:

• To develop an effective control scheme for redundant manipulators using advanced AI.
• To address the inverse kinematics problem in redundant manipulator control.

Main Methods:

• Proposed an online learning fuzzy echo state network (OLFESN).
• OLFESN integrates an online learning echo state network with a Takagi-Sugeno-Kang fuzzy inference system (FIS).
• Devised an OLFESN-based control scheme for redundant manipulators.

Main Results:

• The OLFESN-based control scheme demonstrated effective control of redundant manipulators.
• Simulations and experiments validated the proposed method on UR5 and Franka Emika Panda robots.

Conclusions:

• The OLFESN provides a viable solution for the inverse kinematics of redundant manipulators.
• The proposed control scheme significantly improves the efficiency and instruction of redundant manipulators.