Optimization-based two-dimensional symmetric tossing motion prediction and validation.

Area of Science:

• Biomechanics
• Human Motion Analysis
• Robotics

Background:

• Human motion analysis is crucial for performance and rehabilitation.
• Predicting human motion, especially for injury prevention in repetitive tasks like tossing, is challenging due to the lack of advance experimental data.
• Tossing tasks can lead to significant bodily strain and potential injuries.

Purpose of the Study:

• To present an optimization-based method for predicting 2D symmetric human tossing motion.
• To develop a model that does not rely on pre-collected experimental data.
• To validate the prediction model against experimental data and assess its practical applicability.

Main Methods:

• Utilized sequential quadratic programming to optimize dynamic effort.
• Incorporated both static and dynamic joint torque limits into the optimization.
• Collected experimental data from 10 subjects using motion capture and force plates for validation.

Main Results:

• Predicted optimal tossing motions fell within the experimental standard deviation for kinematic data.
• Predicted ground reaction forces were within the range of experimental data.
• Differences in predicted joint torques were analyzed between dynamic and static strength constraints.

Conclusions:

• The proposed optimization-based method accurately predicts 2D symmetric tossing motion.
• The model's predictions are validated by experimental data, demonstrating its effectiveness.
• Findings have practical implications for injury prevention in occupational settings involving repetitive tossing tasks.