Gotowa bibliografia na temat „Motion consensus inversion”

SUMMARY We investigated the likelihood of radial anisotropy in the shallow and deep upper mantle, including the mantle transition zone (MTZ) under the Indian Ocean. Seismic anisotropy can be an indicator of mantle deformation through lattice preferred orientation of anisotropic crystals in the mantle. It has thus the potential to illuminate Earth’s dynamic interior, but previous seismic tomography studies have not achieved consensus on the existence of radial anisotropy below ∼250 km depth. We developed a fully nonlinear transdimensional hierarchical Bayesian Markov Chain Monte Carlo approach to invert fundamental and higher mode surface wave dispersion data and applied it to a subset of a global Love and Rayleigh wave data set. We obtained posterior model parameter distributions for shear wave velocity (VS) and radial anisotropy ξ under the Indian Ocean. These posterior model distributions were used to calculate the probability of having radial anisotropy at different depths. We demonstrated that separate inversions of Love and Rayleigh waves yield models compatible with the results of joint inversions within uncertainties. The obtained pattern of VS anomalies agrees with most previous studies. They display negative anomalies along ridges in the uppermost mantle, but those are stronger than for regularized inversions. The Central Indian Ridge and the Southeastern Indian Ridge present velocity anomalies that extend to ∼200 km depth, whereas the Southwestern Indian Ridge seems to have a shallower origin. Weaker, laterally variable velocity perturbations were found at larger depths. The anisotropy models differ more strongly from regularized inversion results and their uncertainties were rather large. We found that anisotropy models from regularized inversions also depend on the chosen parametrization, which is consistent with the existence of a large model null-space. Apart from a fast horizontally polarized shear wave signal in the top 100 km, likely reflecting the horizontal plate motion due to asthenospheric deformation, no clear relation to surface geology was detected in the anisotropy models. We found that, although the anisotropy model uncertainties are rather large, and lateral variations are present, the data generally prefer at least 1 per cent anisotropy in the MTZ with fast vertically polarized shear waves, within errors. Incorporating group velocity data did not help better constrain deep structure by reducing parameter trade-offs. We also tested the effect of prior constraints on the 410- and 660-km topography and found that the undulations of these discontinuities had little effect on the resulting models in our study region.

Category: Sports; Ankle Introduction/Purpose: Ankle sprain is one of the most common sports-related injuries. Treatment of these lesions can be performed conservatively with satisfactory results in up to 80% of cases. The main goals of rehabilitation are the recovery of mobility, strength and balance; but there is no consensus on decision make criteria to return to sport. It is known that the time elapsed from trauma may be insufficient reference and may induce early return, while the individual still does not have symmetric parameters of mobility, strength and proprioception. Purpose: To assess the proportionality of muscle strength between the healthy and rehabilitated sides 16 weeks after ankle sprain related to physical activity and to correlate with its impact on function scores. Methods: 131 individuals, from 18 to 64 years old, were evaluated after rehabilitation of simple sprain. Physical examination was performed with anterior drawer test, measurement of leg circumference by perimetry and by the method of figure in '8'; determination of active and passive range of motion using goniometer; muscle strength conference with hand held dynamometer and finally with the application of function scores. Equality test between two proportions, T-student test, Chi-square test and Pearson correlation were used to determine the significance between the values and correlations found. Results: We identified an average age of 36.1 (+- 2.5) years, 56 (42.70%) women and 75 (57.30%) men. 19.8% of sprains were severe. Limb volume and range of motion evaluations did not show significant differences. The anterior drawer remained positive in 19.1% of the individuals with the highest proportion in severe sprains (36.6%). Force asymmetry for flexion (52.7%), extension (50.4%), eversion (46.6%) and inversion (56.5%) movements above the 10% limit were observed. Impact on function scores showed a moderate correlation (0.59) with the proportionality of ankle extension forces. Conclusion: Individuals adequately rehabilitated after an ankle sprain and considered able to return to sports may present changes anterior drawer test and asymmetries in lower limb muscle strength impacting performance on function scores.

Reference/Citation: Hamstra-Wright KL, Bliven KC, Bay C. Risk factors for medial tibial stress syndrome in physically active individuals such as runners and military personnel: a systematic review and meta-analysis. Br J Sports Med. 2015;49(6):362–369. Clinical Question: What factors put physically active individuals at risk to develop medial tibial stress syndrome (MTSS)? Data Sources: The authors performed a literature search of CINAHL, the Cochrane Central Register of Controlled Trials, EMBASE, and MEDLINE from each database's inception to July 2013. The following key words were used together or in combination: armed forces, athlete, conditioning, disorder predictor, exercise, medial tibial stress syndrome, militaries, MTSS, military, military personnel, physically active, predictor, recruit, risk, risk characteristic, risk factor, run, shin pain, shin splints, and vulnerability factor. Study Selection: Studies were included in this systematic review based on the following criteria: original research that (1) investigated risk factors associated with MTSS, (2) compared physically active individuals with and without MTSS, (3) was printed in English, and (4) was accessible in full text in peer-reviewed journals. Data Extraction: Two authors independently screened titles or abstracts (or both) of studies to identify inclusion criteria and quality. If the article met the inclusion criteria, the authors extracted demographic information, study design and duration, participant selection, MTSS diagnosis, investigated risk factors, mean difference, clinical importance, effect size, odds ratio, and any other data deemed relevant. After the data extraction was complete, the authors compared findings for accuracy and completeness. When the mean and standard deviation of a particular risk factor were reported 3 or more times, that risk factor was included in the meta-analysis. In addition, the methodologic quality was assessed with an adapted checklist developed by previous researchers. The checklist contained 5 categories: study objective, study population, outcome measurements, assessment of the outcome, and analysis and data presentation. Any disagreement between the authors was discussed and resolved by consensus. Main Results: A total of 165 papers were initially identified, and 21 original research studies were included in this systematic review. More than 100 risk factors were identified in the 21 studies. Continuous data were reported 3 or more times for risk factors of body mass index (BMI), navicular drop, ankle plantar-flexion range of motion (ROM), ankle-dorsiflexion ROM, ankle-eversion ROM, ankle-inversion ROM, quadriceps angle, hip internal-rotation ROM, and hip external-rotation ROM. As compared with the control group, significant risk factors for developing MTSS identified in the literature were (1) greater BMI (mean difference [MD] = 0.79, 95% confidence interval [CI] = 0.38, 1.20; P < .001), (2) greater navicular drop (MD = 1.9 mm, 95% CI = 0.54, 1.84 mm; P < .001), (3) greater ankle plantar-flexion ROM (MD = 5.94°, 95% CI = 3.65°, 8.24°; P < .001), and (4) greater hip external-rotation ROM (MD = 3.95°, 95% CI = 1.78°, 6.13°; P < .001). Ankle-dorsiflexion ROM (MD = −0.01°, 95% CI = −0.96, 0.93; P = .98), ankle-eversion ROM (MD = 1.17°, 95% CI = −0.02, 2.36; P = .06), ankle-inversion ROM (MD = 0.98°, 95% CI = −3.11°, 5.07°; P = .64), quadriceps angle (MD = −0.22°, 95% CI = −0.95°, 0.50°; P = .54), and hip internal-rotation ROM (MD = 0.18°, 95% CI = −5.37°, 5.73°; P = .95), were not different between individuals with MTSS and controls. Conclusions: The primary factors that appeared to put a physically active individual at risk for MTSS were increased BMI, increased navicular drop, greater ankle plantar-flexion ROM, and greater hip external-rotation ROM. These primary risk factors can guide health care professionals in the prevention and treatment of MTSS.

Abstract—An overview of the results obtained by foreign seismologists based on the records of Turkish seismic networks AFAD (State Agency for Disaster Management under the Ministry of Internal Affairs) is presented. The sequence of earthquakes began with the M7.8 main shock and includes thousands of aftershocks. The strongest events occurred in the first twelve hours, with the sources of two M7.0+ events located 100 km apart. Earthquakes have caused ground motions that are destructive to structures, the so-called “pulse-like waveforms”, and epicentral distances, as was previously noted, are not a good indicator of attenuation of waves from earthquakes with extended ruptures. The records of stations in the near-fault zones clearly revealed the directivity effects of seismic radiation. The M7.8 earthquake (main shock) was larger than expected in the current tectonic setting. The near-field records traced an early transition to the super-shear (~1.55Vs) rupture propagation on the Narli lateral fault, where the rupture originated and then passed into the East Anatolian fault. The early transition to the super-shear stage obviously contributed to the further propagation of the rupture and the initiation of slips on the East Anatolian fault. A dynamic fracture model has been constructed that matches the various results of inversions obtained by different authors and reveals spatially inhomogeneous rupture propagation velocities. Super-shear velocities exceeding the shear wave velocity Vs are observed along the Narli lateral fault and at the southwestern end of the East Anatolian fault. Since the late 1990s, seismologists have been working on incorporating the rupture directivity effects of extended sources into the probabilistic seismic hazard analysis procedures, but no consensus has been reached so far, and progress in this area can only be expected with the accumulation of a sufficient amount of observational data.

Background: Lateral ankle sprains are one of the most common injuries in indoor and court sports. Self-reports and case studies have indicated that these injuries occur via both contact and noncontact injury mechanisms typically because of excessive inversion in combination with plantarflexion and adduction of the foot. Video-based documentation of the injury mechanism exists, but the number of cases reported in the literature is limited. Purpose: To retrieve and systematically analyze a large number of video-recorded lateral ankle injuries from indoor and court sports, as well as describe the injury mechanism, injury motion, and injury pattern across different sports. Study Design: Cross-sectional study; Level of evidence, 3. Methods: A total of 445 unique video-recorded lateral ankle sprain injuries were retrieved from indoor and court sports of broadcasted levels of competition. The videos were independently analyzed by 2 different reviewers. Outcomes included classification of the injury mechanism according to the International Olympic Committee consensus guidelines, primary and secondary motions of ankle joint distortion, and documentation of the fixation point (fulcrum) around which the foot rotates. Results: Overall, 298 (67%) injuries were direct contact, 113 (25%) were noncontact, and 32 (7%) were indirect contact incidents. Direct contact injuries were especially prevalent in basketball (76%), handball (80%), and volleyball (85%), while noncontact injuries dominated in tennis and badminton (96% vs 95% across both). Inversion (65%) and internal rotation (33%) were the primary distortion motions, with the lateral forefoot (53%) and lateral midfoot (40%) serving as the main fulcrums. Landing on another player's foot was the leading cause of injury (n = 246; 55%), primarily characterized by inversion (79%) around a midfoot fulcrum (54%). The noncontact and indirect landings on floor (n = 144; 33%) were primarily characterized by a distortion around a forefoot fulcrum (69%). Conclusion: Two of 3 ankle sprains from online video platforms were direct contact injuries, with most involving landing on another player's foot. The distortion motion seems to be related to the injury mechanism and the fixation point between the foot and the floor. The injury mechanisms varied greatly between sports, and future studies should clearly differentiate and investigate the specific injury mechanisms.

The word 'nigger' is arguably the most charged epithet in American English; thus it is surprising that this word has been appropriated by some African Americans to refer to themselves. To be precise, the African-American version of this term is not 'nigger' but 'nigga', a word that has, as Geneva Smitherman notes, "a variety of meanings ranging from positive to negative to neutral" (Black Talk 167). Henry Louis Gates, Jr., in his study of African-American literature, provides a theoretical foundation for understanding why some African Americans use this word and how it operates rhetorically. Building on Gates's work, I will argue that the co-optation of the slur often involves a complex of three rhetorical devices that fall under the rubric of an African-American rhetorical strategy called Signifyin(g)—a term that will be discussed at length later. The first of these devices is agnominatio, defined as "the repetition of a word with an alteration of both one letter and a sound" (Gates 46). The second, semantic inversion, is the reversal of the meaning of a term (Holt qtd. in Smitherman, "Chain"). Chiastic slaying, the third rhetorical strategy, is a critique that transforms the status of a group or individual.1 Through these three modes of rhetorical transfiguration, the slur 'nigger' becomes 'nigga' a positive term that carries with it a critique of racism. I will further argue that all of these rhetorical devices operate through a principle I term "semantic looping" in which a new term derives meaning by continual reference to an older, existing term. This principle is a key to understanding how Signifyin(g) works in the appropriation of 'nigger' and helps to reveal how, in the words of Michel Foucault, the appropriation is a culturally rooted form of "reverse discourse" (101). Ultimately, this rhetorical analysis reveals that the African-American usage of 'nigga' is a strategy for asserting the humanity of black people in the face of continuing racism, a strategy that celebrates an anti-assimilationist vision of African-American identity. Foucault has argued that while the naming of oppressed groups by those in power serves as an instrument for oppression, such naming can also engender group identification and resistance to oppression (101). The coining of the word 'homosexual', for example, allowed for the repression of gay people but also allowed homosexuals to organise a gay rights movement using the very terminology utilized to oppress them (Foucault 101). One strategy for resisting hostile slurs like 'queer' or 'nigger' is for the oppressed group to appropriate the name and transform it into a rallying cry or "reverse discourse". An understanding of how 'nigga' operates as a reverse discourse requires a culturally rooted rhetorical analysis of the term. Gates, in The Signifyin(g) Monkey, provides background for such an analysis. Because his project is ultimately to derive an African-American theory of literary criticism, he touches on the appropriation of 'nigger' only briefly, asserting that a "political offensive" was mounted against the term by African-Americans through a black rhetorical strategy called Signifyin(g) (47). Gates, however, does not explain precisely how Signifyin(g) works in this case, except to suggest that it involves agnominatio (46). Thus 'nigger' becomes 'nigga', a word that differs from the racial slur but originates from and recalls it.2 Although Gates's commentary on the appropriation of 'nigger' amounts to little more than a sentence, much of his explication of the term Signifyin(g) implicitly applies to the co-optation of 'nigger'. The rhetorical analysis presented in this paper is a logical extension of Gates's initial linkage of the appropriation of 'nigger' with the rhetorical practice of Signifyin(g). The social baggage attached to 'nigga' assures that every use of the term is double-voiced in the Bakhtinian sense. More precisely, 'nigga' is a Bakhtinian parody of 'nigger'; the new connotation parodies or comments on the original because the new term carries with it the history of its pejorative use as well as the refashioned connotation of defiant group pride.3 This kind of rhetorical turn or critique is an example of the African-American rhetorical practice Gates identifies as Signifyin(g). Pinning down exactly what constitutes Signifyin(g) is difficult. Numerous black language scholars have commented on the expansiveness of the term.4 Gates argues that in its broadest sense, to Signify means to be "figurative," further noting that "to define it in practice is to define it through any number of its embedded tropes" (81).5 For our purposes it can be described as a rhetorical action that indirectly critiques another term or sign by revising it. Gates explains that, fundamentally, this revision and critique involve "repetition, with a signal difference" (51). Gates distinguishes the African-American term, 'Signifyin(g)', from the word 'signifying' by capitalizing the 'S' and bracketing the 'g' (46). It is helpful to think of the former term as 'Signifyin(g) on' (or critiquing) something whereas the latter word 'signifies' (or means) something but does not inherently involve a critique. Thus, to parody the motions of a police officer behind his or her back 'Signifies on' the officer and 'signifies' one's disrespect.6 Signifyin(g) is inherently a counter-puncher's strategy, an act of resistance against an oppressive force. Gates even goes so far as to call it the "slave's trope" (52). In Signifyin(g), the revised term, through its parodic double-voicedness, enters into a semantic loop with the original term; recollection of past oppressive usage must occur to fuel the term's new meaning. Figure 1 - Semantic Loop of Semantic Inversion and Agnominatio This semantic loop recalls what W.E. B. Dubois termed African-American double consciousness, a consciousness that yields him no true self-consciousness, but only lets him see himself through the revelation of the other world. It is a peculiar sensation this double consciousness, this sense of always looking at one's self through the eyes of others, of measuring one's soul by the tape of the world that looks on in amused contempt and pity. (16-17) While 'nigga' recalls how blacks have been measured by the tape of the world, it also defies this estimation through ironic revision of the name. Although Dubois would criticize this pathway through the white term as a road to false consciousness, others might insist that since revision of the white term occurs through distinctly African-American rhetorical strategies, the revision is emblematic of an authentically African-American consciousness—which is a double consciousness. In this view the revision does not attempt to reconcile what DuBois calls the "two unreconciled strivings" of the black person as "an American and a Negro" but instead involves them in an endless interplay (17). The interplay of the two signs sustains an antagonistic stand toward the dominant white community through the polemical comment: "this is how whites see us but we are something more". 'Nigga', then, is "authentically black" speech because it recognizes and maintains the divide between black and white worlds. As Smitherman notes: [e]ncoded within the rhetoric of racial resistance, nigga is used to demarcate (Black) culturally rooted from (white) culturally assimilated African Americans. Niggaz are those Bloods (Blacks) who are down for Blackness and identify with the trials as well as triumphs of the Black experience… ("Chain") The defiance implied by the revision of the white slur is also an assertion of human subjectivity. Gates identifies a parallel strategy in African-American slave narratives. Referring to Frederick Douglass's famous chiasmus—"You have seen how a man became a slave, you will see how a slave became a man."—Gates asserts that "Douglass's major contribution to the slave narrative was to make chiasmus the central trope of slave narration, in which a slave-object writes himself or herself into a human subject through the act of writing" (172). By comparison, through the semantic inversion of 'nigger'/'nigga', dehumanized blacks speak themselves into human subjects through the act of speaking. This transfiguration conforms to what Gates terms "chiastic slaying" (66). His somewhat off-hand phrase is inspired by the African-American use of chiasmus, which is defined as, "a grammatical figure by which the order of words in one of two parallel clauses is inverted in the other" (Oxford English Dictionary qtd. in Grothe). Chiasmus is often represented as an ABBA pattern (so Douglass's chiasmus would be reduced to: (A) man - (B) slave - (B) slave - (A) man). In Gates's usage, chiastic slaying involves repetition and reversal but not necessarily a literal ABBA pattern of chiasmus. In the same vein, 'nigga' is a repetition of 'nigger' that reverses the position of African Americans (from objects to subjects). Analogously, 'nigger to nigga' can be conceived of as the inverted second clause of a chiastic statement like Douglass's 'man - slave - slave - man' in which personhood and agency are re-affirmed. This re-affirmation of humanity implicit in 'nigga' is not likely to be understood by many whites given, as Smitherman notes, that they often fail to recognize the semantic difference between 'nigger' and 'nigga'.7 Since whites are frequently unaware of the Signification of 'nigga', it is impossible for African Americans to kill (i.e. end) the white use of the racist term. In the context of Signification, chiastic slaying does not put an end to the idea Signified upon. In fact, Signification must be activated by what Gates calls the "absent presence" of the original term (48). The critique of racism and assertion of subjectivity implicit in the employment of 'nigga' is not aimed at white people or the elimination of their sign; it is aimed at a black audience that must survive in a continually racist environment. What, then, is the "slaying" of chiastic slaying? It must be seen as a refutation of the original term or sign. In the case of 'nigga', it is a rejection of the dehumanization implied by 'nigger' with the recognition that African Americans will still be continually subjected to this libel despite its refutation. Thus, the chiastic slaying of 'nigger' by 'nigga' requires a continual interplay or semantic loop between the two terms. The context of continuing racism, then, requires 'nigga' to recurrently signify on (i.e. assert the falsity of) the slur. The recurrent Signification can be thought of as a loop inscribed upon the linear chiastic pattern: Figure 2 - Semantic Loop Inscribed on the Chiastic PatternThe context of continuing racism is one factor that accounts for the value of semantic looping in African-American rhetoric. Since the semantic loops of African-American culture draw their strength from the oppression to which they react, they are continually useful. This kind of resistance does not attempt to overcome racism but instead draws African-American attention to it so blacks can survive it. The first step in this survival is to be aware, as DuBois might say, that blacks in America are perceived of as a "problem" (15). The Signification of 'nigga' also "keeps it real", by reminding African Americans of the harsh truth of racism and by continually enacting a refutation of racism through a complex of culturally familiar rhetorical strategies. In this respect, the appropriation of the white slur is, to borrow the words of Foucault, a culturally inspired "reverse discourse" aimed at responding to white oppression. The identification of semantic looping in this case opens up an array of other questions. How does semantic looping function in the appropriation of other epithets by other groups? (A few cases that may be worth investigating in addition to the previously mentioned 'queer', are 'dyke', 'girl'/'grrl' by young feminists and 'anorexia'/'ana' as well as 'bulimia'/'mia' by pro-eating-disorder advocates.) Do the cultural differences of various groups affect how semantic looping operates? What does semantic looping reveal about the struggle over authenticity or identity, especially with respect to gender, class and subculture? And lastly, how do groups respond to re-appropriations by dominant groups? (In particular I am thinking of the increasing use of 'nigga' by white American teenagers.) I hope others will find these questions worth pursuing. Notes 1. While Gates suggests that agnominatio is involved in the co-optation of 'nigger', he does not mention the term 'semantic inversion' at all (although he is obviously aware that Signifyin(g) often involves this rhetorical action). Gates's phrase, chiastic slaying, occurs only in the context of a general discussion of Signifyin(g). See 66 in Gates for his use of chiastic slaying. 2. Other English speakers including Australians and the English may find it difficult to distinguish between these spoken words and 'hear' them both as 'nigguh'. But to those from the United States the distinction is noticeable. 3. Gates identifies Bakhtin's notion of the double voiced word and his concept of narrative parody as relevant to African-American rhetoric. See 50, 110-13 and 131 in Gates. Bakhtin's most comprehensive discussion of double-voiced discourse can be found in 185-186, and 190-99. Bakhtin's distinction between parody and other types of discourse can be found in 193-99. 4. Gates lists the following as providing substantive definitions of Signifyin(g): H. Rap Brown, Roger D. Abrahams, Thomas Kochman, Claudia Mitchell-Kernan, Geneva Smitherman and Ralph Ellison (71). Gates considers Mitchell Kernan's data to be more representative than the others' and even she states that she could not get consensus from her informants regarding Signifyin(g) (Gates 80-81). 5. Gates has identified numerous rhetorical strategies that can be involved in Signifyin(g). See 52 in Gates for a complete list of these tropes. 6. I build on an example from Abrahams who states that "... it is signifying to make fun of a policeman by parodying his motions behind his back..." (52). 7. Smitherman notes that the semantic inversion of 'nigger' (or 'flippin the script' as it is known in the hip-hop world) "... is often misunderstood by European Americans and castigated by some African Americans" (Chain). Smitherman's comment suggests that the ability to discriminate between the two terms (as well as one's comfort level with the usage of 'nigga') is not racially monolithic. Whites who participate in hip-hop culture, for example, are likely to see the distinction between 'nigger' and 'nigga'. Some factors that seem likely to complicate any generalization about understanding and comfort level with 'nigga' are race, affinity for hip-hop, class, age and geographic location. References Abrahams, Roger D. Deep Down in the Jungle: Negro Narrative Folklore from the Streets of Philadelphia. Chicago: Aldine, 1970. Bakhtin, Mikhail. Problems of Dostoevsky's Poetics. Ed. and Trans. Caryl Emerson. Minneapolis: U of Minnesota P, 1999. DuBois, W.E. Burghardt. The Souls of Black Folk. Greenwich: Fawcett, 1961. Foucault, Michel. The History of Sexuality: Volume I: An Introduction. Trans. Robert Hurley. New York: Vintage, 1980. Grothe, Mardy. Chiasmus.com. Online. Internet. 9 Oct. 2001. Available <http://www.chiasmus.com/whatischiasmus.shtml>. Gates, Henry Louis, Jr. The Signifying Monkey: A Theory of African-American Literary Criticism. New York: Oxford U P, 1988. Smitherman, Geneva. "'The Chain Remain the Same'." Journal of Black Studies 28 (1997): n.pag. Online. Academic Search Elite. 10 May 2002. - - -. Black Talk. New York: Houghton Mifflin, 1994. Links http://www.chiasmus.com/whatischiasmus.shtml Citation reference for this article MLA Style Jacobs, Andrew T.. "Appropriating a Slur" M/C: A Journal of Media and Culture 5.4 (2002). [your date of access] < http://www.media-culture.org.au/mc/0208/semantic.php>. Chicago Style Jacobs, Andrew T., "Appropriating a Slur" M/C: A Journal of Media and Culture 5, no. 4 (2002), < http://www.media-culture.org.au/mc/0208/semantic.php> ([your date of access]). APA Style Jacobs, Andrew T.. (2002) Appropriating a Slur. M/C: A Journal of Media and Culture 5(4). < http://www.media-culture.org.au/mc/0208/semantic.php> ([your date of access]).

La capacité d'auto-localisation est essentielle pour les véhicules autonomes, les robots, la réalité mixte et plus généralement les systèmes qui interagissent avec leur environnement. Lorsqu’il n’y a pas de carte disponible, les algorithmes de SLAM (Localisation et Cartographie Simultanées) créent une carte de l'environnement et en même temps y localisent le système. Un capteur populaire est la caméra, qui a l'avantage de fournir passivement une représentation visuelle de l'environnement à faible coût, et donc celui que nous utilisons.Le SLAM en environnement dynamique, ou SLAM Dynamique, est un défi car l'algorithme doit être capable de percevoir en permanence quelles parties de l'image sont fixes par rapport au référentiel souhaité par l'utilisateur, en général le sol. Des problèmes surviennent lorsque les hypothèses sur lesquelles reposent les algorithmes SLAM deviennent invalides. Un cas remarquable est l'inversion de consensus de mouvement : lorsque la majeure partie d'une image est constituée d'objets en mouvement, l'algorithme n'utilise pas le bon référentiel, et échoue. Un autre est le masquage excessif : certains algorithmes SLAM retirent des images -- c'est-à-dire masquent -- tous les objets qui pourraient être dynamiques même s'ils ne sont pas en mouvement, et par conséquent échouent si les images deviennent vides.De façon générale, l'utilisateur peut vouloir utiliser un algorithme SLAM dans un contexte non supporté. En réalité, l'écart entre ce dont l'utilisateur a besoin et ce que font les algorithmes SLAM est significatif dans la recherche SLAM et la cause de problèmes tels que les inversions de consensus, elles-mêmes rarement présentes dans la littérature. Ainsi, au lieu de proposer un SLAM plus général, nous proposons un algorithme SLAM qui s'adapte à de nouveaux environnements grâce à un apprentissage auto-supervisé automatisé : apprendre automatiquement quelles parties d'une scène peuvent être mobiles par rapport au référentiel souhaité par l'utilisateur, et quand elles sont fixées ou non. L'utilisateur fournit des vidéos d’entraînement non annotées et notre algorithme apprend automatiquement ce qu'il faut en faire.Nous présentons d’abord l'état de l'art, les bases de données et les métriques SLAM de référence. En particulier, nous détaillons les défis du SLAM Dynamique et de l'évaluation de la robustesse. Les bases de données et métriques SLAM actuelles font partie des points bloquants, nous proposons donc les nôtres. Dans une deuxième partie, nous explorons les relations entre les points d’intérêt d’une image et les performances du SLAM, et à partir de ce travail, nous présentons un nouvel algorithme de SLAM Dynamique auto-supervisé qui apprend quels objets masquer, en utilisant les outliers SLAM. Les outliers SLAM sont des points d’intérêt rejetés au cours du processus de SLAM : nous avons observé que les outliers sur les objets en mouvement ont des propriétés uniques dans des séquences vidéo faciles et peuvent être utilisés pour apprendre automatiquement à segmenter les objets dynamiques. Enfin, nous présentons une approche auto-supervisée qui apprend quand masquer des objets : SLAM Dynamique avec Masquage Temporel. A partir d'une méthode donnée de masquage d’objet, on apprend automatiquement quand masquer les objets de certaines classes. On annote automatiquement chaque image des séquences d’entraînement avec des décisions de masquage (masquer les objets ou non), puis on apprend les circonstances qui ont mené à ces décisions avec un réseau basé mémoire.Les résultats de cette thèse montrent que le SLAM Dynamique auto-supervisé est une approche viable pour résoudre les inversions de consensus de mouvement. Plus généralement, l'auto-supervision est la clé pour qu'un SLAM s'adapte aux besoins des utilisateurs. Nous avons dépassé l'Etat de l'Art en termes de robustesse, en plus de clarifier des points aveugles de la littérature en termes d’évaluation de la robustesse des algorithmes de SLAM Dynamique
The ability of self-localization is essential for autonomous vehicles, robots, mixed reality and more generally to systems that interact with their environment. When maps are not available, SLAM (Simultaneous Localization and Mapping) algorithms create a map of the environment and at the same time locate the system within it. A popular sensor is the camera, which has the benefit of passively providing a visual representation of the environment at a low cost, and for this reason the one we use in this thesis.SLAM in Dynamic environments, or Dynamic SLAM, is challenging as the algorithm must be able to continuously perceive what parts of the image are fixed with respect to the frame of reference the user wants, usually the ground. Problems arise when assumptions SLAM algorithms rely on become invalid. A remarkable case is the Motion Consensus Inversion (MCI): when most of an image is made of moving objects, the SLAM does not use the correct frame of reference and fails. Another one is excessive masking: some SLAM algorithms remove from images -- ie, mask -- all objects that might be dynamic even if they are not moving, and consequently fail if images become empty.More generally, the user may need to use a SLAM algorithm in an unsupported context. In fact, the gap between what the user needs and what SLAM algorithms do is a blind spot in SLAM research and the cause for issues like motion consensus inversions, which are themselves seldom seen in the literature. Hence, instead of making a more general SLAM algorithm, we propose a SLAM algorithm that adapts to new environments through automated self-supervised training: to automatically learn what parts of a scene may not be fixed with respect to the user's desired frame of reference, and when they are fixed or not. The user provides unlabeled training videos and our SLAM automatically learns what to do to from them.In the first part of this document, we present the State of the Art of algorithms for SLAM and Dynamic SLAM, reference datasets and metrics. We detail the challenges of Dynamic SLAM and robustness evaluation. Current SLAM datasets and metrics are also subject to the user need gap, so we propose our own. Our datasets are the first to explicitly include video sequences with motion consensus inversions or excessive masking and our metric is more general that the usual accuracy metrics, which are misleading in very difficult scenarios.In the second part, we explore the relation between image features and SLAM performance, and from this work we present a novel self-supervised Dynamic SLAM that learns what objects to mask, using SLAM outliers. Outliers are features rejected during the standard SLAM process: we observed that outliers on objects in motion have unique properties in easy dynamic sequences. Thus, we locate dynamic objects using outliers and learn to segment them, so we can mask dynamic objects in sequences of any difficulty at runtime.Finally, we present a self-supervised approach that learns when to mask objects: Dynamic SLAM with Temporal Masking. Leveraging an existing method to mask objects, it automatically learns when to mask objects of certain classes. It automatically annotates every frame of training sequences with masking decisions (to mask objects or not), then learn the circumstances that led to these decisions with a memory-based network. We do not make any geometrical assumption, unlike other SLAM algorithms. Using a memory-based approach prevents at runtime motion consensus inversions and excessive masking, which is hardly possible when relying on geometrical methods.The results of this thesis show that a self-supervised Dynamic SLAM is a viable approach to tackle motion consensus inversions. More generally, self-supervision is the key to have a SLAM adapt to user needs. We surpassed the State of the Art in terms of robustness, in addition to clarifying blind spots of the literature in Dynamic SLAM robustness evaluation

Abstract Many phenomena in acoustically loaded structural vibrations are better understood in the time domain, particularly transient radiation, shock, and problems involving nonlinearities and bulk structural motion. In addition, the geometric complexity of structures of interest drives the analyst toward domain-discretized solution methods, such as finite elements or finite differences, and large numbers of degrees of freedom. In such methods, efficient numerical enforcement of the Sommerfeld radiation condition in the time domain becomes difficult; although a great many methodologies for doing so have been demonstrated, there seems to exist no consensus on the optimal numerical implementation of this boundary condition in the time domain. Here, we present theoretical development of several new boundary operators for conventional finite element codes. Each proceeds from successful domain-discretised, projected field-type harmonic solutions, in contrast to boundary integral equation operators or those derived from algebraic functions. We exploit the separable prolate-spheroidal coordinate system, which is sufficiently general for a large variety of problems of naval interest, to obtain finite element-like operators (matrices) for the boundary points. Use of this coordinate system results in element matrices that can be analytically inverse transformed from the frequency to the time domain, using appropriate approximations, without altering the Hilbert space in which the approximate solution resides. The inverse transformation introduces some additional theoretical issues involving time delays and Stieltjes-type integrals, which are easily resolved. In addition, use of element-like boundary operators does not alter the banded structure of the system matrices, which is of enormous importance for efficient solution of large problems. Results presented here include theoretical derivation of the new “infinite elements”, the approximations for certain problematic frequency-domain terms, resolution of the inversion issues, and element matrices for the boundary operators which introduce no new continuity requirements on the fluid field variable.