The main objective of this work was to develop a high-fidelity, nonlinear simulation of complex flexible structures with active control systems. A multibody system can consist of several flexible articulating components. The system can also have multiple control systems. The main challenges are to efficiently model the flexible nonlinear dynamics of the system and integrate multiple control systems with the model.

A number of prior techniques of modeling the dynamics of multibody structures exist. These range from modeling the system as multiple rigid bodies, to assembling the complete system matrices of a flexible multibody system for solving the equations of motion. The main disadvantages of prior art are lack of high-fidelity modeling or lack of computational efficiency.

SOMBAT is a multibody dynamics and control system simulation tool that provides an integrated software environment to perform kinematic and dynamics analysis of space structures and robotic manipulators, including their control elements. The multibody system can consist of an arbitrary number of rigid and flexible bodies in an open loop topology. The dynamics equations of a given system are generated in a computationally efficient and optimized form in SOMBAT using a symbolic code generator.

Using SOMBAT, the components of a multibody system can be modeled as articulating flexible bodies. Hence, a high-fidelity dynamics model of the system can be developed using this approach. The equations of motion in SOMBAT are generated using a symbolic code generator. The code generator takes advantage of the system configuration variables to simplify the expressions in the equations of motion. Time invariant terms are substituted in the equations. Constant terms are combined, redundant terms are eliminated, and sparse matrix operations are optimally implemented. Hence, the generated equations are in a computationally efficient and optimized form. In addition, an Order(n) algorithm is used to solve the system equations. This solution algorithm is recursive in nature, and is superior to conventional solvers for systems with a large number of degrees of freedom.

The main advantage of SOMBAT is the capability to perform high-fidelity, nonlinear simulations of flexible space structures and associated control systems. Environmental disturbances such as those produced by gravity gradient and aerodynamic drag can also be modeled. SOMBAT also includes several sensor and actuator models.

This work was done by Murugan Subramaniam of Akima Infrastructure Services for Johnson Space Center. MSC-25528-1