Optimal Orbit Determination (OOD) methods have the following characteristics:
Optimal Orbit Determination  

Property  Manifestation in OOD 
Input  Tracking measurements with tracking platform locations, an a priori state estimate (inclusive of orbit estimate, and an a priori state error covariance matrix 
Output  Optimal state estimates and realistic state error covariance matrices 
A priori orbit estimate  Required 
A priori state error covariance matrix  Required 
Methods  Filter Methods are forwardtime recursive sequential
machines consisting of a repeating pattern of:

Time transitions (for both methods)  Dominated most significantly by numerical orbit propagators,
OOD methods are characterized by optimality:

Sources  The search for optimality was begun by Wiener, Kalman, Bucy and others. Wright seeks significant improvements for OD using optimal estimation theory. 
Operationally, OOD enables realtime performance, autonomous measurement processing, minimum orbit error variance estimates, and realtime trajectory accuracy performance assessment.
ODTK 6.5