Optimization-based human motion prediction using an inverse-inverse dynamics technique implemented in the AnyBody Modeling System

This paper presents an optimization-based human movement prediction using the AnyBody modeling system (AMS). It is explained how AMS can enables prediction of a realistic human movement by means of a computationally efficient optimization-based algorithm.

The human motion predicted in AMS is based on a physics model including dynamic effects and a high level of anatomical realism. First, a musculoskeletal model comprising several hundred muscles is built in AMS. The movement is then parameterized by means of time functions controlling selected degrees of freedom of the model. Subsequently, the parameters of these functions are optimized to produce an optimum posture or movement according to a user-defined cost function and constraints. The cost function and the constraints are typically express performance, comfort, injury risk, fatigue, muscle load, joint forces and other physiological properties derived from the detailed musculoskeletal analysis. The technique is demonstrated on a human model pedaling a bicycle. We use a physiology-based cost function expressing the mean square of all muscle activities over the cycle to predict a realistic motion pattern.

Posture and motion prediction are important because they enhance the usability of digital manikins for different purposes. In the field of product design for instance, movement prediction techniques entail the possibility for product designers to reduce the number of time consuming and costly experiments. From an orthopedics point-of-view, human movement simulation enables the experts to predict the effect of prosthetic and orthotic devices.