Based on the artificial immune system paradigm and a hierarchical multi-self strategy, a set of algorithms for aircraft sub-systems failure detection, identification, evaluation and flight envelope estimation have been developed and implemented. Data from a six degrees-of-freedom flight simulator were used to define a large set of 2-dimensional self/non-self projections as well as for the generation of antibodies and identifiers designated for health assessment of an aircraft under upset conditions. The methodology presented in this paper classifies and quantifies the type and severity of a broad number of aircraft actuators, sensors, engine and structural component failures. In addition, the impact of these upset conditions on the flight envelope is estimated using nominal test data. Based on immune negative and positive selection mechanisms, a heuristic selection of sub-selves and the formulation of a mapping-based algorithm capable of selectively capturing the dynamic fingerprint of upset conditions is implemented. The performance of the approach is assessed in terms of detection and identification rates, false alarms, and correct prediction of flight envelope reduction with respect to specific states.