Dynamic Human Model EMMA: Software for Simulating Human Movements

Muscle Cords or Muscle Torque Generators

There are two modeling approaches to choose from: On the one hand, muscles can be represented as individual strands that are connected to bone segments at several points. Each strand represents the properties of a real muscle. On the other hand, so-called Muscle Torque Generators (MTG) can be used. They combine the effect of all muscle segments that move a joint in a certain direction. This approach significantly reduces the complexity of the model without sacrificing a realistic representation – only biarticular muscles, i.e. those that act across two joints, are not yet taken into account.

Dynamics of Muscle Activation Included

EMMA also takes into account the temporal dynamics of muscle activation: The software simulates how activation can change within a very short time and thus approximates the process in which neuronal signals are converted into muscle strength. This ensures a realistic representation of force development – from signal transmission to movement execution.

Active Motion Generation: Should Emma Act or React?

EMMA calculates dynamic movements using modern optimization methods. In doing so, the software minimizes a cost function and simultaneously fulfils all boundary and constraint conditions defined by the user. The activation of the muscle model is calculated automatically - and can be checked for overloads afterwards.

Whether ergonomic evaluation, biomechanical analysis or simulation of complex movement sequences - EMMA offers numerous functions for creating and researching a wide range of movement tasks. Custom models, for example for objects or environments with which the human model is to interact, can be easily integrated as a multi-body system. A few examples can be found in the video.

Two Approaches for Movement Prediction

EMMA provides two methodological approaches for the prediction of human movement:

1. Anticipatory Behavior: Act Optimally
   In this variant, the human model knows all future events or there are no unforeseen disruptions. The software calculates an optimal movement based on a defined cost function – such as minimum effort or shortest execution time. This method is particularly suitable for trained movement sequences such as work processes or sports routines.
2. Responsive Behavior: Reacting Flexibly to Disruptions 
   If external influences such as unexpected forces or accelerations occur, the human model must react to them. In this case, EMMA uses a method that decouples the motion planning from the future state of the system. This results in responsive, adaptive behavior – for example, when simulating occupants in vehicles or when dealing with unknown technical systems.