New Neural Network for Real-Time Human Dynamic Motion Prediction

Research in the field of human simulation has led to significant advancement in quality, time, and cost management for products like military and athletic equipment and vehicles. There is, however, a critical need for human simulation models to run in real time, especially those with large-scale problems like motion prediction (a single motion problem involves prediction of between 500-700 outputs). Hence, this work addresses that need by developing a new design of artificial neural network (ANN) that is capable of providing real-time motion results with maximum accuracy and minimal training. The success of the new ANN design is proven for the intended problem of motion simulation and other experimental and real-world problems. In addition, the design creates a new tool for the analysis of the task being simulated. The new implemented ANN algorithms will open a new area of advancement and capability in the digital human modeling field. Although the new ANN design is driven by its use with motion prediction problems, the consequent ANN design can be used with a broad range of large-scale problems. The motion problem is simply a well-studied example for the proposed developments

Autorentext
Dr. Bataineh received his Ph.D. in Bioengineering from the University of Iowa. He currently works as data scientist in clinical analytics at Humana Inc. He conducts research in machine learning for clinical applications. His research interest includes data mining, machine learning,neural networks, digital human modeling, and biomechanics