The dynamic model used for the Ranger NBV flight simulator is
almost identical to the one presented in Section 4.1.1. The only
difference is that buoyancy offset has not yet been added to the
simulation.
The values of the simulator physical parameters were determined
experimentally by comparing simulated and actual vehicle angular
acceleration and maximum velocity about each axis, and iterating
until the simulation matched the actual system performance.
The simulator models thruster maximum capacity (or saturation). The
simulator is normalized to use a maximum torque of 1 Nm about each
axis. Assuming (incorrectly) this torque level for Ranger NBV yields
the inertia matrix used by the simulator.

_{} kgm^2 
(4.58) 
The off diagonal elements are zero indicating that Ranger NBV is
symmetric about the vehicle axes. These values can be modified to
model asymmetries caused by various manipulator configurations.
Combining (4.58) with the thruster performance data from Table 27
gives an estimate of the actual underwater inertia matrix. The
underwater inertia matrix is different than the inertia observed on
the surface because it also includes the virtual mass created by the
water inside of and surrounding the vehicle.

_{} kgm^2 
(4.59) 
The normalized values of the drag matrix used in the simulation are

_{} kgm^2. 
(4.60) 
These values were determined by observing the terminal angular
velocity about each of the vehicle axes. The values chosen in the
drag matrix approximate the vehicle with the manipulators stowed.
This symmetric configuration causes the off diagonal elements to be
zero.
The normalized values in (4.60) are matched to the simulator which
has a maximum moment capability of 1 Nm about each axis. Again using
the propulsion system performance data we find an approximation of
the actual drag of Ranger NBV. After continued testing the drag
figures have been determined to be about 30% to 50% too high.

_{} kgm^2 
(4.61) 