Rozprawy doktorskie na temat „Terminal guidance”

Thesis (M.S. in Electrical Engineering) Naval Postgraduate School, March 2000.
Thesis advisor(s): Hutchins, Robert G. "March 2000." Includes bibliographical references (p. 83). Also available online.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 2008.
Includes bibliographical references (leaves 117-121).
Small munition effectiveness is a function of miss distance from the target and ability to achieve a steep flight path angle at the target. Many small guided munitions have limited control authority to achieve these competing objectives due to system hardware tradeoffs. This thesis develops guidance algorithm modifications that demonstrate consistent improvement in achieving these objectives over previously used methods with changes only to the flight software and not the hardware or system concept of operations. Most modifications attempt to intelligently incorporate post-launch information into the guidance system, however there is an investigation into better using pre-launch information through dynamic programming. Dynamic programming is an off-line approach to optimize the guidance parameters applied in flight, based on measurable flight characteristics. All investigated methods demonstrate varying abilities to improve performance for this munition system. While dynamic programming is computationally intensive, it produces an efficient look up table which is easily implemented in real time with minimal additional memory requirements. The thesis further shows that performance improvements are gained by altering the rocket ignition time in flight, by tightening the tolerances on some key sources of modeling error, and by developing a highly accurate time to impact estimation algorithm. Regardless of the particular modification, better utilizing pre- and post-launch information improves the munition's performance and utility for the user. While not tested, simultaneously implementing several of these improvements could further increase performance.
by Paul C. Tisa.
S.M.

A unique feature in the topographic organization of the olfactory bulb (OB) is the dual role of the odorant receptor (OR). It detects odors but it also plays an instructive role in the convergence of olfactory sensory neuron axons expressing the same odorant receptor to form glomeruli in specific loci of each olfactory bulb. This segregation of sensory afferents gives rise to the olfactory map, that has a critical role in encoding odors. How the OR could accomplish such different roles has remained enigmatic for more than 20 years. If the OR can determine axon convergence, it should act as an axon guidance molecule. The demonstration that the OR is expressed not only at the cilia, but also at the axon terminal, a suitable location for an axon guidance molecule, corroborated this hypothesis. However the mechanism of activation and the function of the OR at the axon terminal has remained unknown for all these years. The odorant receptors (OR) expressed at the cilia, is a G-protein-coupled receptor, that upon binding odorant molecules, activates a specific G-protein, Golf, that stimulates adenylyl cyclase type III (ACIII) to produce cyclic AMP (cAMP). cAMP, in turn, opens olfactory specific nucleotide-gated (CNG) channels, driving an influx of Ca2+ and Na+ within the cell. The rise of intracellular calcium concentration opens Ca2+ activated Cl channels, leading to a Cl- efflux, that further depolarizes the membrane potential. More recently, Lodovichi's group demonstrated for the first time that the odorant receptor expressed at the axon terminus-growth cone of the olfactory sensory neurons is functional and coupled to local increase of cAMP and Ca2+. These second messengers play a key role in axon elongation and turning in several systems, including the olfactory system, where it has been shown that the OR-derived cAMP contributes to the location of the glomeruli in the olfactory bulb. The open critical question that remained to be addressed was the mechanism of activation and function of the odorant receptor at the axon terminal. In particular, what are natural ligands of the odorant receptor expressed at the axon terminal? A few molecules expressed in the olfactory bulb could bind and activate the odorant receptors expressed at the axon terminal. To verify this hypothesis, Lodovichi's group purified several groups of molecules (by chromatographic procedures) from the embryonic rat olfactory bulb and tested their ability to activate the OR at the axon terminal. Performing real time Ca2+ imaging in olfactory sensory neurons, they observed that a pool of molecules from the olfactory bulb, locally applied, was able to elicit a prompt Ca2+ rise at the axon terminal. By mass spectrometry of the active pool of molecules, phosphatidylethanolamine-binding protein-1 (PEBP1), also known as Raf kinase inhibitory protein (RKIP), was identified as a putative ligand of the axonal OR. PEBP1 is a ~21 KDa protein that belongs to a highly conserved family of proteins, expressed in numerous tissues and cell types in a variety of species. In rodents, PEBP1 is expressed in several brain areas, both in neurons and in non-neuronal cells. The physiological function of PEBP1 in the brain has remained obscure. Its low molecular weight, its ability to be secreted, to modulate G-protein coupled receptor (GPCR) and the presence of olfactory deficits in mice carrying a null mutation for PEBP1 (PEBP1-/- mice), make PEBP1 a putative OR ligand. During my PhD work, I have studied whether PEBP1 could activate the OR and its contribution in the formation of the sensory map. First, we studied Ca2+ dynamics in response to PEBP1, in rat and mouse olfactory sensory neurons (OSN) loaded with the calcium indicator, fura-2, and transfected with a specific odorant receptors (OREG, P2, ORS6, Olfr62 and M72) We found that rat embryonic olfactory sensory neurons exhibited a prompt Ca2+ rise in response to PEBP1 locally applied at the axon terminal with a glass pipette. Noteworthy, PEBP1 extracted from rat bulb elicited a prompt Ca2+ rise also at axon terminus of mouse olfactory sensory neurons, suggesting that this response is highly conserved between species. To ascertain that the Ca2+ rise in response to PEBP1 observed in OSN was due to OR receptor activation, we performed Ca2+ imaging in Human embryonic kidney 293 cells (HEK293Tcells) transfected with specific odorant receptors (OREG, P2, S6, Olfr62, M72), loaded with the calcium indicator fura-2 and challenged with PEBP1 and with the specific odor ligand (used as control). HEK293Tcells transfected with a specific odorant receptor exhibited a prompt Ca2+ rise in response to PEBP1 and to the cognate odor for the specific odor receptor HEK293Tcells transfected with the odorant receptor M72 did not exhibit Ca2+rise in response to PEBP1 but only to the M72 cognate-odor ligand methylsalicylate. This result indicates that there are likely a few other ligands that can activate the OR. Indeed, HEK293Tcells transfected only with the empty-vector (pCI, not expressing any odorant receptors) did not exhibit Ca2+rise in response to PEBP1, nor to the cognate odor ligand. These data demonstrated that PEBP1 elicits Ca2+response via odorant receptor activation. HEK293T cells loaded with fura-2 and transfected with specific OR, did not present a Ca2+ response when challenged with PEBP1 denatured by higher temperature or with proteinase K. To exclude that possible contaminants in the purification procedure, could trigger Ca2+ responses, HEK cells expressing specific OR and loaded with fura-2, were challenged with a cyclin-dependent kinase 2 (CDK2), a protein isolated with the same purification procedure of PEBP1. No calcium rise was observed in response to CDK-2, confirming the specificity of PEBP1 response. Axon guidance molecules steer the direction of axon elongation. To explore whether that was the case for PEBP1, we performed time-lapse imaging of single olfactory sensory neuron axon behaviour in response to a gradient of molecules able to modulate cAMP and Ca2+ signaling at the axon terminal (i.e forskolin). We found that PEBP1 regulates the turning behaviour of sensory axons with a similar effect of forskolin (FRSK) or odors able to modulate the level of cAMP and Ca2+ at the growth-cone of OSN. The turning behaviour of OSN was not affected by Ringer’ solution, used as a negative control. Where is PEBP1 expressed in the olfactory bulb? By immunofluorescence on embryonic, postnatal rat and mouse olfactory bulb (OB) sections, we found that PEBP1 is expressed in periglomerular cells, a suitable location to act on incoming axons. PEBP1 is expressed with a patchy pattern mostly in the anterior (A), medial (M) and lateral (L) side of the bulb while in the posterior (P) side is hardly detected. By immunofluorescence, RT-PCR and western blot we observed that PEBP1 was not expressed in olfactory sensory neurons (OSNs) located in the main olfactory epithelium (MOE) of embryonic, postnatal rats. Accordingly, no staining for PEBP1 was observed in the core of the glomeruli. For all these techniques, we used as a negative control tissues from mice carrying a null mutant of PEBP1 (PEBP1-/- mice). If PEBP1 plays an important role in the 1). organization and 2). location of glomeruli in the OB, then PEBP1 mutant mice (PEBP1-/- mice), should exhibit a disrupted sensory map. To verify this hypothesis PEBP1 mutant mice (PEBP1-/- mice), were crossed with mice in which specific OR is co-expressed with GFP (such P2-GFP , M72–YFP mice) and we analyze the convergence of GFP positive fibers to form glomeruli in the bulb. We found that in PEBP1 -/-mice the convergence of sensory neurons expressing P2-OR were disrupted by the presence of several additional heterogeneous glomeruli while sensory neurons expressing M72 (not responsive to PEBP1 in vitro) were not affected. Noteworthy, the alteration in the topographic organization of the sensory map in PEBP1 -/- mice crossed with P2-GFP mice was stronger in homozygous than in heterozygous mice, showing a dosage dependent effect. We then evaluated the location of specific glomeruli in the OB, a crucial aspect of the topographic map. We computed the position of the main P2 and M72-homogeneous glomeruli (i.e. formed exclusively by sensory axons expressing the same OR) in whole-mount bulbs along the antero-posterior (A-P) and dorso-ventral axis (D-V). We found that the position of the P2-glomeruli was significantly shifted along the antero-posterior axis (A-P) in PEBP1 mutant mice compared to controls. The defect was stronger in homozygous than in heterozygous PEBP1-/- mice crossed with P2-GFP mice, showing a dosage dependent effect. No defects were reported along the D-V axis, in agreement with the presence of specific molecules that dictate the location of glomeruli along the D-V axis. The location of M72-glomeruli was similar in control and PEBP1-/-, in agreement with the unaltered organization of the M72 glomeruli and with our in vitro data, in which the odorant receptor P2, but no M72, was responsive to PEBP1. To explore the functional outcome of the altered sensory map, we analyzed the olfactory behaviour in PEBP1-/- mice. We performed an olfactory habituation-dishabituation test that showed an olfactory deficit in the detection and discrimination of a new odor in PEBP1-/- mice compared to control mice. The data obtained in my PhD thesis suggested that the OR at the axon terminal can act as an axon guidance molecule activated by the natural ligand identified, PEBP1.
Una caratteristica peculiare dell'organizzazione topografica del bulbo olfattivo (OB) è il "duplice ruolo" dei recettori olfattivi (OR). Esso, infatti, non è unicamente coinvolto nella percezione degli odori, ma svolge anche un ruolo cruciale nella convergenza assonale di neuroni sensoriali olfattivi (OSN) esprimenti lo stesso OR, nella formazione di glomeruli in specifiche regioni del OB. Tale organizzazione delle afferenze sensoriali dà origine alla mappa sensoriale. Ciò nonostante, il meccanismo attraverso cui l'OR svolga queste due distinte funzioni è rimasto sconosciuto per più di 20 anni. Se l'OR svolgesse uan funzione nella convergenza assonale, potrebbe agire come "axon guidance", molecole guida, per gli assoni stessi. A sostegno di questa ipotesi, è stato dimostrato che l'OR è espresso non solo alle cilia del neurone sensoriale olfattivo, ma anche nell'assone terminale, assumendo quindi una posizione ottimale per guidarne il movimento. Il loro meccanismo di attivazione, così come la loro funzione all'assone terminale, rimane però sconosciuta. I recettori olfattivi (OR) espressi nelle ciglia, sono recettori accoppiati a proteine G che, in seguito al legame con molecole odorose, attivano una specifica proteina G, Golf, che stimola l'adenilato ciclasi di tipo III (ACIII) a produrre AMP ciclico (cAMP). Quest'ultimo induce l'apertura di specifici canali regolati da nucleotidi ciclici chiamati "cyclic nucleotide-gated (CNG)", con conseguente influsso di ioni Ca2+e Na+ nel citoplasma della cellula. Tale aumento intracellulare di Ca2+ attiva canali di Cl- che, permettendo l'uscita di ioni Cl- dal citoplasma, depolarizzando la cellula. Recentemente, il gruppo della Dott.ssa Lodovichi ha dimostrato per la prima volta che il recettore olfattivo espresso all'assone terminale (cono di crescita) dei OSN è funzionale e accoppiato ad un locale incremento di cAMP e Ca2+. Ca2+ e cAMP sono due secondi messaggeri che svolgono un ruolo chiave nell'allungamento e nel "turning" movimento assonale in diversi sistemi, compreso il sistema olfattivo, in cui è stato dimostrato che cAMP che si produce in seguito all'attivazione del recettore olfattivo contribuisca a formare i glomeruli in una specifica posizione nell' OB. Il punto critico che rimane da affrontare e definire è il meccanismo di attivazione e la funzione del recettore olfattivo espresso all'assone terminale dei OSN. In particolare: Quali siano i ligandi naturali espressi all'assone terminale? Un gruppo ristretto di molecole espresse nel bulbo potrebbero legare e attivare il recettore olfattivo espresso all'assone terminale. Al fine di verificare questa ipotesi il gruppo della Dottoressa Lodovichi è stato in grado di identificare un gruppo di molecole (mediante tecniche cromatografiche) estratte da bulbi olfatti di ratti embrionali e ne ha testato la capacità di attivare l'OR all'assone terminale. Attraverso la tecnica dell'imaging del calcio in neuroni sensoriali olfattivi, il gruppo della Dott.ssa Lodovichi ha osservato che un gruppo di molecole applicate localmente, è in grado di attivare i neuroni sensoriali olfattivi, determinando una risposta di calcio all'assone terminale. Mediante spettrometria di massa delle molecole in grado di attivare l'OR è stato identificato un possibile ligando dell'OR espresso all'assone terminale dei OSN. Si tratta di una proteina, apparentemente alla famiglia delle etanolammine, chiamato phosphatidylethanolamine-binding protein-1 (PEBP1) o inibitore delle Raf chinasi (RKIP). PEBP1 è una proteina di circa ~21 KDa che appartiene ad una famiglia di proteine molto conservata espressa in diversi tessuti di diverse specie. Nei roditori PEBP1 è espressa in diverse aree cerebrali, sia in neuroni che in cellule non neuronali. La funzione fisiologica di PEBP1 nel cervello rimane tuttavia non nota . Il suo basso peso molecolare, il fatto che venga secreta, che moduli i recettori accoppiati a proteine G, nonchè la presenza di deficit olfattivi in linee di topo che possiedono mutazioni di PEBP1 (topi PEBP1 -/-), rendono PEBP1 un possibile ligando del recettore olfattivo. Durante il mio lavoro di tesi di dottorato ho investigato se e in che modo PEBP1 potesse attivare l'OR e il suo contributo nella formazione della mappa sensoriale olfattiva. Come primo step, sono state studiate le dinamiche del calcio in risposta a PEBP1 sia nei neuroni sensoriali di ratto che di topo, caricati con l'indicatore di calcio, fura-2. Questi esperimenti hanno permesso di identificare risposte calcio in seguito all'applicazione di PEBP1 all'assone terminale di OSN di ratto embrionale mediante una pipetta di vetro. Degno di nota è che PEBP1, estratto dal bulbo olfattivo di ratto è in grado di indurre risposte calcio anche in OSN di topo. Questa evidenza suggerisce che l'aumento di calcio in risposta all'applicazione di PEBP1 è altamente conservata tra le specie. Al fine di verificare che l'aumento di calcio in risposta di PEBP1 osservato in OSN fosse dovuto all'attivazione di OR, esperimenti di calcium imaging sono stati effettuati in linee cellulari derivate da cellule endoteliari di rene embrionale umano (HEK293T). Alle linee cellulari HEK, trasfettate con uno specifico OR (OREG, P2, S6, Olfr62, M72) caricate con l'indicatore per il calcio fura-2, è stato applicato PEBP1 e il corrispondente odore del OR trasfettato. Nelle linee cellulari HEK293T trasfettate con specifici OR (OREG, P2, S6, Olfr62) è stata osservata un'immediata risposta di calcio intracellulare in seguito sia all'applicazione di PEBP1 che dell'odore specifico del OR trasfettato usato come controllo); Un' eccezione è stata riscontrata in HEK293T esprimenti il recettore M72 che sono responsive soltanto per l'odore specifico di M72 (Methylsalicilate) ma non a PEBP1. Questo risultato suggerisce la presenza di altre molecole, oltre che a PEBP1, che possano attivare l'OR. Inoltre, in HEK293T trasfettate con il solo vettore pCI, non esprimente l' OR, non è stato osservato alcun aumento di calcio intracellulare nè in risposta a PEBP1, nè all'odore specifico del OR trasfettato, ma solamente in risposta al carbacolo, usato come test di vitalità cellulare. Questi risultati mettono in evidenza che l'aumento di calcio intracellulare in seguito all'applicazione di PEBP1 è dovuto all'attivazione del OR. Tuttavia, PEBP1 denaturato mediante alte temperature o con proteinase K non determina alcuna risposta di calcio intracellulare in HEK293T trasfettate con OR specifico. Al fine di escludere che la presenza di contaminanti nella procedura di purificazione, possa indurre una risposta calcio nelle linee cellulari HEK esprimenti uno specifico OR è stato applicata un'altra proteina chinasi 2 ciclina-dipendente (CDK2), purificata con lo stesso protocollo di PEBP1. Nessuna risposta di calcio intracellulare è stata osservata confermando la specificità di risposta di PEBP1. Molecole guida "axon guidance" dirigono l'allungamento assonale. Al fine di investigare se PEBP1 fosse in grado di modulare il movimento assonale, è stato valutato il comportamento del cono di crescita mediante la tecnica di " time-lapse imaging"sul singolo assone in risposata a gradienti di molecole in grado di modulare il signaling dei secondi messaggeri cAMP e Ca2+ all'assone terminale (es. foscolina). Abbiamo osservato che PEBP1 regola il movimento dell'assone di OSN con un simile effetto della foscolina o degli odori, capaci di modulare i livelli di cAMP e Ca2+ al cono di crescita del neurone sensoriale olfattivo. Dov'è espresso PEBP1 nel bulbo olfattivo? Mediante la tecnica di immunofluorescenza su sezioni di OB di ratto embrionale, postanatale e di topo, è stata osservata l'espressione di PEBP1 nelle cellule periglomerulari, un' ottima posizione per guidare gli assoni degli OSN. PEBP1 è espresso con un pattern discontinuo, maggiormente nella regione anteriore (A), mediale (M) e laterale (L) dell'OB mentre è difficile osservare cellule positive per PEBP1 nella zona posteriore (P). Mediante immunofluorescenza, RT-PCR e western blot è stato osservata l'assenza di espressione di PEBP1 nei OSN localizzati nell'epitelio olfattivo (MOE) di ratti embrionali, postnatali e di topo. In accordo con queste evidenze, PEBP1 non è espresso nel "cuore" dei glomeruli. Per tutte queste tecniche, tessuti provenienti da una linea di topo con mutazioni per PEBP1 (topi PEBP1-/-) sono stati utilizzati come controlli negativi. Se PEBP1 giocasse un ruolo importante nell' organizzazione e nella posizione dei glomeruli nel bulbo olfattivo (OB), linee di topi che presentano mutazioni per questa proteina (topi PEBP1-/-), dovrebbero mostrare alterazioni nella mappa sensoriale. Al fine di verificare questa ipotesi, i topi che presentano mutazioni per PEBP1 (topi PEBP1-/-) sono stati incrociati per linee di topi in cui uno specifico OR è co-espresso con una proteina fluorescente, GFP (es. come le linee di topi P2-GFP , M72–YFP). In queste linee di topi è stata studiata la convergenza delle fibre positive per la proteina fluorescente (GFP) nella formazione dei glomeruli nell'OB. La convergenza assonale dei OSN esprimenti il recettore P2-OR è distrutta dalla presenza di numerosi glomeruli eterogenei, mentre quella delle fibre esprimenti il recettore M72 (non responsivo a PEBP1 in vitro) non è alterata. Degno di nota è che l'alterazione nell'organizzazione topografica della mappa sensoriale è maggiormente evidente nei topi omozigoti per la mutazione di PEBP1 rispetto ai topi eterozigoti, suggerendo un meccanismo dosaggio dipendente. Successivamente, è stata valutata la posizione di specifici glomeruli nel OB, un punto cruciale della mappa topografica. Nessun difetto è stato riportato lungo l'asse dorso-ventrale (D-V) del bulbo, in accordo con la presenza di altre molecole che guidano le afferenze assonali dei OSN nella formazione dei glomeruli in una specifica posizione lungo l'asse dorso-ventrale del OB. La posizione del glomerulo M72 non differisce tra i topi PEBP1-/- e controlli. Questo dato è in accordo con una normale organizzazione dei glomeruli in cui convergono le fibre-M72 e con i dati in vitro in cui il recettore P2 ma non M72 è responsivo per PEBP1. Al fine di investigare l'effetto funzionale dell'alterazione della mappa sensoriale è stato effettuato un test olfattivo (habituation-dishabituation test) nei topi PEBP1-/-. E' stato riscontrato un deficit olfattivo nel riconoscimento e discriminazione degli odori nei topi PEBP1-/- rispetto ai controlli. Tutti i dati raccolti in questa tesi di Dottorato suggeriscono che OR espressso all'assone terminale può agire come molecola guida attivate dal ligando naturale del OR identificato PEBP1.

The human protein Brahma (Brm), encoded by the SMARCA2 gene, is one of the two mutually exclusive ATPase subunits of the mammalian SWI/SNF-BAF chromatin-remodelling complex. Brm-containing BAF complexes are enriched in neurons, where they play crucial roles in the regulation of genes involved in neuronal differentiation. Moreover, it has been reported that Brm associates with components of the spliceosome to regulate the inclusion of alternative internal exons. While investigating with splicing-sensitive microarrays the gene expression changes triggered by mitochondrial stress, I found that Brm is strongly downregulated in SH-SY5Y human neuroblastoma cells overexpressing the SOD1 (G93A) protein, one of the genetic causes of Amyotrophic Lateral Sclerosis (ALS). I found that this downregulation is due to a mitochondrial stress-induced impairment in the SMARCA2 promoter activity. Among the genes deregulated at the splicing level by SOD1 (G93A) expression, I identified several targets that are regulated by alternative 3’ terminal exon usage in a Brm-dependent manner. Specifically, I found that Brm promotes the skipping of the proximal terminal exon in five out of six genes that were analyzed. In order to define the molecular mechanism that allow to Brm to modulate the choice of alternative 3’ terminal exons, I used one of these genes, RPRD1A, as a model. I found that Brm inhibits the choice of the proximal RPRD1A last exon by directly localizing in its genomic region. In turn, the absence of Brm at the level of the proximal last exon is concomitant with a change in the processivity of the RNA Polymerase II, an observation consistent with the “terminal exon pausing” event. I hypothesized a model where Brahma may recruit the Bard1-Cstf complex on the RPRD1A proximal last exon, a complex known to inhibit the 3’ end processing of the pre-mRNA. These observations suggest an inhibitory role for Brm, which is exerted both at the level of the cotranscriptional choice of the proximal last exon and at the level of the 3’ end pre-mRNA processing.

C-terminal binding proteins (CtBPs) are recruited by a variety of transcription factors to mediate gene repression. C. elegans CTBP-1 has previously been shown to play a role in the regulation of lifespan; Caenorhabditis elegans strains carrying a deletion in the ctbp-1 gene showed a 10-20% increase in mean and maximal lifespan compared with wild-type control strains. We set out to identify the tissues in which CTBP-1 functions to regulate lifespan in C. elegans. Our analysis of reporter genes shows that CTBP-1 is predominantly expressed in the nervous system with lower levels detectable in the hypodermis. Tissue-specific rescue experiments demonstrated that CTBP-1 functions in the nervous system to regulate lifespan. Previously, the lifespan extension in a ctbp-1 mutant was attributed, at least in part, to the misregulation of a lipase gene, lips-7. We found that expressing CTBP-1 solely in the nervous system of a ctbp-1 mutant significantly reduced lips-7 transcription. Locomotory behaviours such as exploration behaviour are regulated by a neuronal circuit involving many subtypes of neurons and incorrect development or function of these neurons results in changes in these behaviours. We observed reduced exploration behaviour in ctbp-1 mutants. Our examination of a subset of neurons involved in regulating locomotion, among which are the dorsal SMD (SMDD) neurons, revealed that SMDD axons are misguided in ctbp-1 mutants. Expressing CTBP-1 under the control of the endogenous ctbp-1 promoter rescued the SMDD axon guidance defect in ctbp-1 mutants. Additional rescue experiments suggest that CTBP-1 functions in the nervous system to regulate SMDD axon guidance. Interestingly, the pre-synaptic marker RAB-3 was found to localise to the misguided portion of SMDD axons in a ctbp-1 mutant. In summary, our findings show that CTBP-1 can function in the nervous system to regulate longevity, gene expression and axon guidance.

碩士
國立交通大學
電機與控制工程系
88
In this thesis, we study the interception problem of homing missiles. From the robust stabilization point of view, we propose two of robust control techniques and use them on the design of missile guidance law. When the missile pursuits the target, the acceleration of target is usually unknown and can be regarded as external disturbance. According to the robustness of sliding-mode control, we use Variable Structure Control scheme to design missile guidance law. Moreover, H-infinity control scheme is also proposed for missile interception. We wish the ratio of output energy and input energy will less than or equal to the prescribed value. We use the capture time and the cumulative velocity increment as two performance indices. For the purpose of performance comparison, we compare the performance indices of two missile guidance laws with those of TPN and IPN. From simulation results, we can find that VSC can achieve good performance against both non-maneuvering and maneuvering target and the performance can be controlled by the parameters chosen by the designer.

In the guidance literature, minimizing miss distance along with optimizing the energy usage had been an objective for several decades. In current day applications, additional terminal performance such as impact angle and impact time are of paramount importance. These terminal constraints increase warhead effectiveness and survivability of the interceptor. This thesis contributes to the design of guidance laws addressing terminal constraints such as impact angle, impact time, and both impact time as well as impact angle, in addition to interception of targets. In the first part of the thesis, the guidance laws which ensure the alignment of the interceptor at a desired impact angle within a finite time is proposed using different variants of sliding mode control(SMC).The impact angle is first redefined in terms of line-of-sight angle and then the impact angle problem is converted to a simpler problem of controlling line-of-sight angle and their rates. The sliding mode capturability and interpretation of the guidance laws are presented. In order to cater to very large heading angle errors, which give rise to negative closing speed initially, modifications to the guidance laws are also suggested. The modifications to the guidance laws for avoiding singularities, which may be encountered during implementation, due to the inherent nature of terminal SMC, are suggested. However, the guidance laws, which alleviates the possibility of such singularities completely, are also designed by using non singular terminal SMC. The two loop guidance and control, for a skid-to-turn cruciform interceptor in the pitch plane, is also proposed with an autopilot designed using the concept of dynamic SMC. The guidance laws addressing impact angle constraint for three dimensional scenarios are also presented. Unlike the usual approach of decoupling the three dimensional engagement in to two mutually orthogonal planar engagements, the guidance laws are derived using coupled engagement dynamics. These guidance laws are designed using conventional and non singular terminal SMC and provide asymptotic and finite time alignment of the intercept or to the desired impact angles, respectively. Next, the SMC based guidance laws which ensure the interception of targets at pre-specified impact times is proposed in this thesis. The guidance law is first designed for stationary targets and then extended to constant velocity targets using the notion of predicted interception point. A switching surface is designed using the concepts of collision course and time-to-go with non-linear engagement dynamics and its role in achieving the objectives is also discussed. In order to account for large heading angle errors and even for negative initial closing speeds, different methods of estimation of time-to-go, resulting in two different guidance laws, are used. Unlike the existing guidance laws, the proposed guidance laws achieve an impact time even less than its initially estimated value. The flexibility in selecting a desired impact time is also exploited using the maximum available acceleration information. A cooperative salvo attack strategy, based on the proposed impact time guidance law, with a desired impact time chosen in real time using a centralized coordination algorithm, is proposed for stationary targets. The coordination manager determines a common impact time based on time-to-goof the interceptors, by minimizing the total switching surface deviations which in turn reduces the control effort. The thesis also proposes a SMC based guidance strategy which addresses impact angle and impact time constraints simultaneously. This guidance scheme is based on switching between impact time and impact angle guidance laws based on certain conditions. Unlike existing impact time guidance laws, the proposed guidance strategy takes into account the curvature of the trajectory due to the impact angle requirement. The interceptor first corrects its course to nullify the impact time error and then aims to achieve interception with desired impact angle. In order to reduce the transitions between the two guidance laws, a novel hysteresis loop is introduced in the switching conditions. Initially stationary targets are considered, and later the same guidance scheme is extended to constant velocity targets using the notion of predicted interception point. Theclaimsofalltheguidancelawsarevalidatedwithextensivesimulationsandtheir performances are compared with existing guidance laws. Although all the guidance laws derived in the thesis are based on the assumption of constant speed interceptors, their performances are evaluated with a time-varying speed interceptor model, subjected to aerodynamic conditions, to validate their efficacy. The implementation of impact time guidance on time-varying speed interceptors is a formidable challenge in the guidance literature. Such implementations have also been presented in the thesis after introducing the notion of average speed and shown to yield satisfactory performance.

碩士
國立臺灣大學
電機工程學系研究所
86
The styles of war progress with the development of science. To attack the enemy by launching ballistic missiles must come true in the future. To defense the threat of ballistic missiles is a very important work in many advanced countries. Some topics about the terminal guidance problems of high altitude anti-ballistic-missile missile are studied in this thesis, from the references to the theater-high-altitude-area-defense (THAAD) project of U.S. Army. The imaginary target, which the intercept missile must impact, is with the advantage of velocity (about twice of the velocity of the missile). Also with the limitation of the structure of the missile and the flight environment, the missile can only output thrust with fixed amplitude and perpendicular to the axis of the missile. The divert thrust is used to change the flight trajectory of the missile. A kind of proportional navigation guidance law based on the uniform concept of zero-effort-miss (ZEM) is proposed, and it is used to compute the acceleration command according to the attitude of the missile. Also, the guidance law can estimate the initial conditions the midcourse guidance must achieve. To decide the firing of impulse thrust and torque, a new firing logic based on the prediction of ZEM is proposed, too. The divert control system and the attitude control system are integrated and the performance is demonstrated by numerical simulation.

碩士
國立臺灣大學
電機工程研究所
84
This thesis is an integration of two earlier results on mis- sile guidance and autopilot, and is devoted to intercept maneuv- erable, destructive missiles with high speeds. The guidance design is a nonlinear implementation of the co- nstant bearing guidance. This guidance algorithm is intended to control the pursuer in a collision course to hit the target. Its linear implementations such as proportional navigation (PN) and its improved variants are the most widely applied guidance algo- rithms nowadays. The guidance command in the form of acceleration is transfo- rmed into acceptable forms for the autopilot design. This design incorporates feedforward neural networks in the structure of ad- aptive control. Its weighting matrices and other unknown parame- ters are tuned on-line, and the approximation error of the neur- al networks is compensated via the technique of sliding mode co- ntrol. It is proved to be capable of controlling the nonlinear, uncertain and time varying missile system. Extensive simulations are performed to simulate the interce- pt processes and to verify the validity of the overall system.

Current renewed interest in exploration of the moon, Mars, and other planetary objects is driving technology development in many fields of space system design. In particular, there is a desire to land both robotic and human missions on the moon and elsewhere. The core of a successful landing is a robust guidance, navigation, and control system (GN&C). In particular, the landing guidance system must be able to deliver the vehicle from an orbit above the planet to a desired soft landing, while meeting several constraints necessary for the safety of the vehicle. In addition, due to the performance limitations of current launch vehicles, it is desired to minimize the amount of propellant used during the landing. To make matters even more complicated, the landing site may change in real-time in order to avoid previously undetected hazards which become apparent during the landing maneuver. The Apollo program relied heavily on the eyes of the astronauts to avoid such hazards through manual control. However, for missions to the lunar polar regions, poor lighting conditions will make this much more difficult; for robotic missions, this is not an option. It is desired to find a solution to the landing problem such that the fuel used is minimized while meeting constraints on the initial state, final state, bounded thrust acceleration magnitude, and bounded pitch attitude. With the assumptions of constant gravity and negligible atmosphere, the form of the optimal steering law is found, and the equations of motion are integrated analytically, resulting in a system of five equations in five unknowns. When the pitch over constraint is ignored, it is shown that this system of equations can be reduced analytically to two equations in two unknowns. In addition, when an assumption of a constant thrust acceleration magnitude is made, this system can be reduced further to one equation in one unknown. It is shown that these unknowns can be bounded analytically. An algorithm is developed to quickly and reliably solve the resulting one-dimensional bounded search. The algorithm is used as a real-time guidance and is applied to lunar and Mars landing test cases.
text

The next generation of reusable launch vehicles (RLVs) require significant improvements in guidance methods in order to reduce cost, increase safety and flexibility, whilst allowing for possible autonomous operation. Research has focused on the ascent and hypersonic re-entry flight phases. This thesis presents a new method for trajectory design, optimisation and guidance of RLVs during the terminal area flight phases. The terminal area flight phase is the transitional phase from hypersonic re-entry to the approach and landing phase. The trajectory design, optimisation and guidance methods within this thesis are an evolution of previous work conducted on the ascent and re-entry flight phases of RLVs. The methods are modified to incorporate the terminal area flight phase through the adaption of the problem definition and the inclusion of the speed brake setting as a steering parameter. The methods discussed and developed in this thesis are different to previous methods for the terminal area flight phase as they encompass optimisation, trajectory design and guidance based on the definition of the steering parameters. The NLPQL nonlinear optimiser contained within the International Mathematics Standards Library (IMSL) is utilised for trajectory design and optimisation. Real-time vehicle guidance is achieved using the restoration steps of an accelerated Gradient Projection Algorithm (GPA). The methods used are evaluated in a three degrees of freedom (3DOF) simulation environment. To properly evaluate the programs and gain a better understanding of the terminal area flight phase, two different vehicles are utilised within this study. These vehicles are the German sub-orbital Hopper concept vehicle, a previously proposed replacement for the Ariane series of launch vehicles and the recently cancelled joint National Aeronautics and Space Administration (NASA) and Lockheed Martin sub-orbital test bed vehicle, X-33. The two vehicles each have a terminal area flight phase, but their mission profiles and vehicle characteristics are significantly different. The Hopper vehicle is a winged re-entry vehicle, whereas the X-33 vehicle is a lifting body. The trajectory design method takes into account the initial and final conditions, in-flight restrictions such as dynamic pressure and vehicle loads as well as safety margins. The designed trajectories are evaluated to analyse the terminal area flight phase and to assist in the development of the guidance program. The guidance method is evaluated utilising an program consisting of two parts, a real world simulator with high order models and a representation of the on-board guidance computer, the predictor, which uses low order models for computational efficiency. The guidance method is evaluated against a variety of off-nominal conditions to account for dispersions within the high order real world models and common errors experienced by re-entry vehicles. These off-nominal conditions include atmospheric disturbances, winds, aerodynamic, mass, navigation, steering and initial condition errors. The results of this study include a detailed analysis of the terminal area flight phase highlighting the major influences for vehicle and trajectory design. The study also confirms the applicability of the non-linear programming method utilising the vehicle steering parameters as a viable option for trajectory design and guidance. A comparison to other available results highlights the strengths and weaknesses of the proposed method.
http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1282342
Thesis (Ph.D.)--School of Mechanical Engineering, 2007.

The next generation of reusable launch vehicles (RLVs) require significant improvements in guidance methods in order to reduce cost, increase safety and flexibility, whilst allowing for possible autonomous operation. Research has focused on the ascent and hypersonic re-entry flight phases. This thesis presents a new method for trajectory design, optimisation and guidance of RLVs during the terminal area flight phases. The terminal area flight phase is the transitional phase from hypersonic re-entry to the approach and landing phase. The trajectory design, optimisation and guidance methods within this thesis are an evolution of previous work conducted on the ascent and re-entry flight phases of RLVs. The methods are modified to incorporate the terminal area flight phase through the adaption of the problem definition and the inclusion of the speed brake setting as a steering parameter. The methods discussed and developed in this thesis are different to previous methods for the terminal area flight phase as they encompass optimisation, trajectory design and guidance based on the definition of the steering parameters. The NLPQL nonlinear optimiser contained within the International Mathematics Standards Library (IMSL) is utilised for trajectory design and optimisation. Real-time vehicle guidance is achieved using the restoration steps of an accelerated Gradient Projection Algorithm (GPA). The methods used are evaluated in a three degrees of freedom (3DOF) simulation environment. To properly evaluate the programs and gain a better understanding of the terminal area flight phase, two different vehicles are utilised within this study. These vehicles are the German sub-orbital Hopper concept vehicle, a previously proposed replacement for the Ariane series of launch vehicles and the recently cancelled joint National Aeronautics and Space Administration (NASA) and Lockheed Martin sub-orbital test bed vehicle, X-33. The two vehicles each have a terminal area flight phase, but their mission profiles and vehicle characteristics are significantly different. The Hopper vehicle is a winged re-entry vehicle, whereas the X-33 vehicle is a lifting body. The trajectory design method takes into account the initial and final conditions, in-flight restrictions such as dynamic pressure and vehicle loads as well as safety margins. The designed trajectories are evaluated to analyse the terminal area flight phase and to assist in the development of the guidance program. The guidance method is evaluated utilising an program consisting of two parts, a real world simulator with high order models and a representation of the on-board guidance computer, the predictor, which uses low order models for computational efficiency. The guidance method is evaluated against a variety of off-nominal conditions to account for dispersions within the high order real world models and common errors experienced by re-entry vehicles. These off-nominal conditions include atmospheric disturbances, winds, aerodynamic, mass, navigation, steering and initial condition errors. The results of this study include a detailed analysis of the terminal area flight phase highlighting the major influences for vehicle and trajectory design. The study also confirms the applicability of the non-linear programming method utilising the vehicle steering parameters as a viable option for trajectory design and guidance. A comparison to other available results highlights the strengths and weaknesses of the proposed method.
Thesis (Ph.D.) -- University of Adelaide, School of Mechanical Engineering, 2007.

碩士
國立臺北科技大學
機電整合研究所
101
Issues regarding the design of midcourse guidance laws for antimissiles are addressed. The antimissile is expected to be guided to a place with a desired direction, where a ballistic missile is pre- dicted to pass in the reverse direction, so that the target can be easily found and locked for terminal in- terception. The predicted location and direction of a ballistic missile may vary with time, due to infor- mation update using a trajectory prediction algorithm. To fulfill the guidance performance, the guid- ance laws are designed by combining the Takagi-Sugeno (T-S) fuzzy approach and the Terminal Sliding Mode Control (TSMC) technique. Under the designed guidance law, it is shown that the antimissile is able to be efficiently guided to a specified location and direction, even when the existence of uncertainties and disturbances.

Text in German
Forschungsgegenstand dieser Arbeit ist das Bewältigungsverhalten von schwer kranken und sterbenden Menschen, die in einer Gesellschaft ihren bevorstehenden Tod bewältigen müssen, die ambivalent mit dieser Thematik umgeht. Die unterschiedlichen menschlichen Bewältigungsformen können mit den bisher gängigen Zugangswegen über Phasenmodelle nicht immer befriedigend erklärt werden. Hilfreich erscheint hier das aus der Stressforschung stammende Copingmodell von Richard S. Lazarus und seiner Forschungsgruppe, um Menschen verstehen zu lernen. In wieweit dieses Verstehensmodell aus einem säkularen Forschungszweig für die Seelsorgearbeit unterstützend sein kann, soll in dieser Arbeit geklärt werden. Als Grundlage dienen Seelsorgegespräche mit unterschiedlichen Menschen, die eine große Bandbreite an Bewältigungsverhalten demonstrierten. Die Herausforderung lag darin, Menschen in ihrem Sterben umfassend wahrzunehmen, um möglichst viele Impulse zu erkennen, die ihr Bewältigungsverhalten beeinflussten. Aus der Aufarbeitung dieses Verhaltens ergaben sich Perspektiven für die Begleitung sterbender Menschen. Dies hat gleichbedeutend Konsequenzen für den seelsorgerlichen Umgang mit Sterbenden. Das Wissen um das christlich-biblische Menschenbild und die umfassende Kenntnis der Copingforschung können sich durchaus gegenseitig befruchten, denn beide gehen auf die Individualität des Menschen ein und unterstützen ihn seinen eigenen Zugang zu seinem Leben und seinem Tod finden zu lassen. Dies führt letztlich zu einer intensiven Kommunikation zwischen Patient und Seelsorger, der den Mut und das Vertrauen haben darf, dem Sterbenden eigenständig sein Leben bewältigen zu lassen. Die Ansätze der Copingforschung bieten weitreichende Impulse für vielfältige seelsorgerliche Fragestellungen auch außerhalb der Sterbebegleitung. Object of research of this thesis is the specific behaviour of seriously ill and dying humans coping with their impending death and being members of a society having an ambivalent attitude towards death. People's different and individual ways of coping cannot always be sufficiently explained by the current use of phasing models. In this context the "Coping Model" by Richard S. Lazarus and his stress research group seems to be helpful in understanding dying humans. In how far this comprehension model from a secular branch of research can be supportive of pastoral counselling is the objective of this thesis. Counselling dialogues with different persons who show a wide range of coping form the basis of this research. The challenge of this paper was to observe and sense humans in their death experiences from a broader point of view in order to identify as much as possible kinds of impulses crucial to their coping behaviour. Analysing the behaviour of dying humans led to new perspectives on spiritual guidance. Tantamountly, this has its consequences for spiritual dealing with dying humans. The comprehension of the biblical Christian idea of man and a detailed knowledge of coping research can, by all means, be interactive and thus show positive results as both respond to a person's individuality and support them in their own ways of approach to life and death. Finally, this leads to an intensive dialogue between patient and pastor who can have the courage and the confidence in letting the person find his own way of coping. The different kinds of approach of the coping model research offer far-reaching impulses to various pastoral questions going even beyond terminal care.
Philosophy, Practical & Systematic Theology
D.Th. (Practical Theology)