Dynamic modelling for implementation of a right turn in bipedal walking

Area of Science:

• Robotics
• Biomechanics
• Control Systems

Background:

• Controlling the dynamic motion of biped robots, especially during complex maneuvers like turning, remains a significant challenge in robotics.
• Existing models often simplify muscle actuation or lack comprehensive control over ground reaction forces, limiting realistic simulation and application.
• Understanding the interplay between skeletal dynamics, muscle forces, and trajectory control is crucial for developing advanced bipedal locomotion.

Purpose of the Study:

• To develop a conceptual nonlinear control model for a seven-link biped robot performing a turning maneuver.
• To derive the equations of motion for the bipedal skeletal model.
• To compute the necessary muscular forces for executing a desired turning trajectory and controlling ground reaction forces.

Main Methods:

• Derivation of equations of motion for a seven-link bipedal skeletal model.
• Idealization of lower limb muscles as simple force actuators, excluding agonist-antagonist co-contraction.
• Implementation of a nonlinear control scheme using desired trajectories and ground reaction force patterns as input.

Main Results:

• Successful derivation of the biped's equations of motion.
• A nonlinear control strategy was proposed and validated using input data from existing literature.
• The control strategy effectively computed the required muscular forces for the turning maneuver.

Conclusions:

• The developed conceptual model and nonlinear control scheme are effective for controlling a multilink biped during turning.
• The approach allows for the computation of muscular forces necessary to achieve desired trajectories and manage ground reaction forces.
• This work provides a foundation for more sophisticated bipedal robot control systems.