|
The engagment of penaided, nuclear armed Ballistic Missile Re-entry Vehicles (RVs) by a Theatre Missile Defence (TMD) system requires the use of a robust and adaptive discriminsation system to identify warheads from accompanying decoys and other penetration aids. TMD systems will be characterised by their electronic countermeasure environments, and short flight times of the ballistic missile threat. In such environments time is of the essence for TMD commanders to make effective decisions about the allocation of defence weapon systems. The identification and classification, i.e the discrimination, of warheads in a theatre environment is therefore especially stressing requiring detailed analysis and quantification. The discrimination system must be capable of processing object-attribute data derived by sensor systems throughout the four major flight regimes (boost, post boost, mid course, terminal) of ballistic missiles, in many different measurement domains, i.e in multi-feature space. Historically discrimination system designs have been based on a deterministic approach where the defence assumes that signature differences will eventually appear between the warheads and the penetration aids. The classic example is the use of the K factor, where Gaussian distributions - with different mean values for the warheads and penetration aids - are assumed for specific feature space. These classification methodologies process the multi-spectral and multi-spatial data using algorithms designed to separate features within these domains. This allows, in the first instance, the identification of different groups of threat objects. This is followed by their classification into either warheads or one of a number of penetration aids - perhaps through some form of population analysis. These methods rely upon identifiable differences appearing between warheads and penetration aids and are therefore known to be vulnerable to robust countermeasure designs. This has lead to a penaid matching philosophy based on minimisation of the Bhattacharyya Distance. The approach adopted by the research described within this paper is to add inferencing algorithms to the discrimination process - in combination with a database of a-priori information on likely countermeasure options, and their associated signatures. The discrimination system is then able to adapt to the measurements made during the course of the raid, and reason about the types and numbers of countermeasure employed by the offence; through this process postulating the configuration of the penetration aids on the missile systems. In this way specific features of interest can rapidly be identified allowing the discrimination system to concentrate on those areas in preference to those where no worthwhile data is available; thus optimising the use of sensors' viewing time in the course of battle. The paper reviews the principles behind the use of an Artificial Intelligence (AI) based approach to discrimination, and describes the design of an evaluation facility established to quantify the utility of this methodology.
|
