Research was conducted to determine the feasibility of adaptively identifying the values of the hydrodynamic parameters of a submarine or model from submerged free-flight trajectory data. The modified Kalman filter approach which was used was based on the equation-error parameter identification principle instead of the more conventional response error method, resulting in a relatively modest computational requirement. If the controls are driven by appropriate pseudo-random noise sequences, it is possible to determine almost all the parameters to good accuracy from a five minute trajectory sample which is contaminated by realistic additive noise plus bias. Several configurations of suitable instruments for measuring the trajectory are described and the algorithms for processing the trajectory data are discussed. (Author).