Research On The Target Motion Analysis With Bearings-only And Intelligent Guidance Laws Of Torpedo

The guidance is the essence of torpedo effectively attacking target, and the rigorous target motion parameters is the major premise of the guidance. Studying the target motion parameters and the torpedo guidance law have a profoundly influence on torpedo fulfilling it's tactics task. Observation platform, torpedo and target as investigated object, this dissertation deeply investigates the correlation between observation platform and target and the correlation between torpedo and target. After acquiring the target parameters via target motions analysis with bearings-only based on maximum likelihood estimation and Kalman filtering, the torpedo perpendicular impact guidance law based on genetic algorithm and neural network are studied in terms of artificial technology dealing with non-linearity question. The main contributions of the dissertation are summarized as follows:1. Observability criteria for target motion analysis with bearings-only(1) An explicit expression of observability criteria for target motion analysis with bearings-only is derived using classic theory. The similarity between it and the Nardone and Aidala' criteria is proved by means of elementary algebra.(2) The properties of derived criteria are developed, and the relevant proof is given using exterior algebra. The practical applications are given under many conditions.2.Target motion analysis (TMA) with bearings-only(1) TMA with bearings-only based on MLE with batch processing is deeply investigated. The methods of TMA are studied under three circumstances: single observation platform, double observation platforms and three dimensions. The respective simulation is performed, and the affect factors on algorithms are also comprehensive analyzed.(2) The recursive extended Kalman filtering of TMA in modified polar co-ordinates is developed. The corresponding simulations are studied under two cases: evasive target and non-evasive target. The fusion algorithm of distributed Kalman filtering based on multiple hydrophones is proposed.(3) The basic scheme of TMA system is constructed. Subsequently, the system software is compiled and basic interface is also produced.3.TMA with bearings-only using multiple observation platforms(1) The model of TMA with bearings-only using multiple observation platforms is established. The measurements of TMA with bearings-only using multiple observation platforms are studied.(2) The observability of TMA with bearings-only using multiple observationplatforms is given under two scenarios: maneuvering target and nonmaneuvering target. The three theorems on observability are derived and detailed analyzed using exterior algebra.(3) The properties of TMA with bearings-only using two observation platforms are thoroughly studied. The state estimation variances is qualitatively investigated under three cases, the approximation explicitly expression of variances subsequently given under certain azimuth conditions.(4) In TMA with bearings-only using multiple observation platforms multiple, the explicit expression of state estimation variances is obtained through Fisher information matrix (FIM) expanded, and comparison with the Weinstein' conclusion for short integration times is also carried.4.Perpendicular impact guidance law based on genetic algorithm(1) The many appeared torpedo wired-guided guidance law are systematically summarized, the mathematical model and analytical expression are also derived.(2) The torpedo homing guidance and wake guidance laws are concluded, and the principle, characteristic of many algorithms are compared.(3) A gradual angle of lead torpedo homing guidance law based on genetic algorithm is proposed, and the corresponding simulation is also carried.5. Perpendicular impact guidance law based on neural network algorithm(1) The theory of Perpendicular impact guidance law based on neural network algorithm is deeply studied; the feasibility and superiority of neural network in torpedo guidance are analyzed according to the...