Optimization-based subject-specific planar human vertical jumping prediction: Model development and validation

Abstract

Jumping biomechanics may differ between individuals participating in various sports. Jumping motion can be divided into different phases for research purposes when seeking to understand performance, injury risk, or both. Experimental-based methods are used to study different jumping situations for their capabilities of testing other conditions intended to improve performance or further prevent injuries. External loading training is commonly used to simulate jumping performance improvement. This paper presents the optimization-based subject-specific planar human vertical jumping to develop the prediction model with and without a weighted vest and validate it through experiments. The skeletal model replicates the human motion for jumping (weighting, unweighting, breaking, propulsion) in the sagittal plane considering four different loading conditions (0% and 10% body mass): unloaded, split-loaded, front-loaded, and back-loaded. The multi-objective optimization problem is solved using MATLAB® with 35 design variables and 197 nonlinear constraints. Results show that the model is computationally efficient, and the predicted jumping motion matches the experimental data trend. The simulation model can predict vertical jumping motion and can test the effect of different loading conditions with weighted vests and arm-swing strategy on the ground reaction forces. This work is novel in the sense that it can predict ground reaction forces, joints angles, and center of mass position without any experimental data.