Neural Network Based Contact Force Control Algorithm for Walking Robots

Area of Science:

• Robotics
• Control Systems
• Artificial Intelligence

Background:

• Classical contact force control for walking robots requires precise models or sensors.
• Existing methods face limitations in efficiency and disturbance rejection.

Discussion:

• A novel neural network-based contact force control algorithm is proposed.
• The algorithm integrates with a linear quadratic regulator for position control and balance.
• It accurately generates force and reduces errors without external sensors.

Key Insights:

• The neural network model achieves high accuracy in force generation for walking robots.
• It significantly outperforms Jacobian-based calculations in force control accuracy.
• Simulations confirm the algorithm's robustness and rapid disturbance rejection capabilities.

Outlook:

• This approach offers a sensor-independent method for precise robotic locomotion.
• It paves the way for more efficient and stable legged robots.
• Further research can explore real-world implementation and complex terrains.