The advent of democrary in South Africa in the early 1990's led, due to choice or circumstances, to the influx of numerous learners whose home language is an indigenous language, into school environments where the language of learning and teaching (LoLT) is English. These learners are confronted with the challenge of mastering the curriculum content while simultaneously learning English in which it is couched and instructed. When the LoLT differs from learners’ home language, this difference can be an educational barrier, which if not addressed timeously and effectively, may become a major challenge to a fully inclusive education system. The main aim of this research was to investigate whether Grade 4 English second language learners who experienced barriers to English as LoLT were supported in all subjects. The research also probed how effectively these learning support strategies were being implemented under the guidance of SBSTs at mainstream schools. / Inclusive Education / M. Ed. (Inclusive Education)

The advent of democrary in South Africa in the early 1990's led, due to choice or circumstances, to the influx of numerous learners whose home language is an indigenous language, into school environments where the language of learning and teaching (LoLT) is English. These learners are confronted with the challenge of mastering the curriculum content while simultaneously learning English in which it is couched and instructed. When the LoLT differs from learners’ home language, this difference can be an educational barrier, which if not addressed timeously and effectively, may become a major challenge to a fully inclusive education system. The main aim of this research was to investigate whether Grade 4 English second language learners who experienced barriers to English as LoLT were supported in all subjects. The research also probed how effectively these learning support strategies were being implemented under the guidance of SBSTs at mainstream schools. / Inclusive Education / M. Ed. (Inclusive Education)

La problématique de cette thèse réside dans la caractérisation et le maintien de la stabilité des Véhicules Légers Tout Terrain (VLTT). Elle se concentre plus particulièrement sur le développement de systèmes de sécurité actifs capables à la fois de prévenir le conducteur des risques encourus mais aussi de les limiter afin d'assurer l'évolution du véhicule dans une zone de stabilité prédéfinie. Comme le cadre expérimental privilégié est l'application à la stabilité des quadricycles légers à moteurs, plus connus sous le terme anglophone "quad", une des contraintes du projet a été de se limiter à un système sensoriel bas-coût afin d'être en mesure d'industrialiser un tel système. En premier lieu, les métriques de stabilité (Transfert de Charge Latéral et Longitudinal : TCLa et TCLo) ont été choisies grâce à une étude préliminaire sur la stabilité des VLTT. Par la suite, une modélisation 2D en roulis et en tangage avec la prise en compte des déplacements du pilote sur le véhicule sont présentées, ce qui permet d'estimer respectivement le TCLa et le TCLo uniquement à partir de la mesure de l'accélération latérale et longitudinale. Étant donné que pour la suite des travaux, l'anticipation du risque de renversement latéral est nécessaire, un modèle 2D en lacet du véhicule est proposé afin d'obtenir un modèle analytique décrivant la dynamique latérale du véhicule. La suite du mémoire présente les différentes techniques d'observation proposées pour l'estimation des variables et paramètres non-directement mesurables du modèle en lacet du véhicule et qui influencent sa stabilité latérale : les glissements, les conditions d'adhérence et les inclinaisons du véhicule. Plusieurs observateurs ont été proposés, dont le dernier permet de considérer des conditions d'adhérence différentes entre les essieux avant et arrière en utilisant plus largement les accélérations mesurées. Cela permet d'intégrer les passages de sous- à sur-vireur qu'il est essentiel de considérer quand on étudie la stabilité de ce type de véhicule. Ainsi, l'estimation des glissements est toujours pertinente, ce qui permet d'obtenir par la suite une meilleure prédiction de la métrique de stabilité latérale (TCLa) quel que soit le comportement du véhicule. Puis en s'appuyant sur les estimations des observateurs couplées aux modèles dynamiques du véhicule et sur l'extrapolation des commandes du conducteur sur un horizon de prédiction, il est possible de prédire les évolutions du TCLa. Cette valeur prédite ainsi que les estimations en ligne des métriques de stabilité constituent alors le point d'entrée pour la synthétisation d'un système de sécurité actif dédié aux VLTT. Celui-ci est basé sur la génération d'un retour d'effort au niveau de la gâchette des gaz permettant soit d'informer le pilote du risque encouru par la création d'une sensation de dureté, soit d'imposer le retour complet de la gâchette des gaz, ce qui implique une diminution de la vitesse et donc la réduction du risque. Finalement, dans le cas où il est possible de maîtriser la vitesse du véhicule par l'installation d'un système de rétroaction sur les freins (Quad haut de gamme ou robot mobile), les derniers travaux présentés s'intéressent aux techniques de commande prédictive à modèle afin de calculer en temps-réel la vitesse maximale admissible, qui assure l'évolution du critère de stabilité choisi dans un domaine de stabilité. Les modèles, les observateurs, la prédiction du TCLa et les 2 systèmes de prévention présentés dans ce mémoire ont été validés et testés au travers de simulations avancées et d'essais expérimentaux réalisés sur un quad agricole et un robot autonome. Il apparaît alors qu'en plus d'être efficace pour la prévention des risques de renversement à hautes dynamiques, le système de sécurité est industriellement viable. Cela a été rendu possible grâce à une conception reposant uniquement sur des actionneurs et un système sensoriel, dont les coûts sont en adéquation avec le prix d'un VLTT. / This dissertation addresses the topic of lateral dynamic stability of light All-Terrain Vehicles (ATV) in a o-road environment. In particular, the developments are focused on the design of active security system allowing both to prevent the driver of the rollover risk and to ensure the vehicle stability within a safety margin. As the experimental framework is focused on light ATV (e.g. quad bikes), the main limitation of the project has been to keep a low cost sensing equipment in order to keep a possible industrialization. First, the stability metrics (Lateral and Longitudinal Load Transfer : LaLT and LoLT) have been chosen thanks to a preliminary study on the ATV stability. Then, two 2D models (roll and pitch) are proposed allowing to estimate the LaLT and the LoLT from the only acceleration measures and within the driver displacements on the vehicle. However, as the anticipation of lateral risk is mandatory for the next stage of the work, a 2D yaw model is also proposed in order to obtain an analytic model of the vehicle lateral dynamic. Then, the different observation techniques are presented to estimate the variables and parameters of the yaw model, who are not directly measurable and who inuence the lateral stability : sliding, grip conditions and vehicle slopes. Several observers are proposed, whose the last one permits to estimate separately the grip conditions between the front and rear axle thanks to a particular use of the acceleration measure. It allows to take into account the switch between the under- and over-steering behaviour influencing largely the vehicle stability. Thus, the sliding estimations are always relevant what permits to obtain a better prediction of the lateral stability metric (LaTT) whatever the vehicle behaviour. Then, thanks to the observer estimations combined with the vehicle dynamic models and an extrapolation of the driver inputs on a prediction horizon, the Lateral Load Transfer can be predicted. This latter value as well as the online estimations of the two stability metrics are the input of the active security system, which is based on the generation of a counter-force on the trigger. Thus, this system allows first to inform continuously the driver of the incurred risk thanks to a trigger feel harder to press. Secondly, if the risk is too important, the system is able to force the full return of the trigger, what reduces the vehicle speed and consequently the risk. Finally, a control law based on the Predictive Functional Control (PFC) theory has been designed to ensure the lateral stability of the vehicle. This control law allows to calculate in real time the maximal speed ensuring the evolution of the stability metric into a safety range. Nevertheless, this solution is only available for vehicles where it is possible to regulate their speed thanks to a control system on the brake. The models, the observers, the prediction of the LaLT and the two prevention systems presented in this dissertation have been validated and tested through several advanced simulations, a actual quad bike and autonomous mobile robot. Moreover, the security system appears to be directly integrated to industry, as its design rests upon low cost actuators and sensors compared to the price of actual quad bikes.

This study investigated teachers' and learners' perceptions of the causes of poor performance in Mathematics among Grade 9 learners in Durban's Ethusini Circuit of KwaZulu-Natal Province. By using a phenomenographic research design, a sample of 15 Grade 9 learners who take Mathematics, one Grade 9 Mathematics teacher and one Mathematics head of department was selected purposively from three high schools in the Ethusini Circuit. The three schools are one former model C school, one ordinary government school and one private school. Focus group and semi-structured interviews were used to collect data from the participants. The data were analysed by transcribing the focus group and semi-structured interviews; thereafter, the transcriptions were coded into themes and categories. The results of this study indicate that teachers and learners perceive that poor performance in Mathematics in Grade 9 is due to CAPS curriculum challenges, such as overloaded and advanced content, and insufficient time to complete the curriculum. Parental support, English (first additional language) as a language of learning and teaching, methods of teaching (such as the talk and chalk method), and learners' negative attitude towards Mathematics were also cited. From these findings, recommendations for revisiting the CAPS curriculum and other areas of need, as well as topics for future research were suggested. / Dinyakišišo tše di nyakišišitše ka ga maikutlo a barutiši le a baithuti ka ga dilo tšeo di bakago go se šome gabotse ka thutong ya Dipalo gareng ga baithuti ba Kreiti ya 9 ka Sedikothutong sa Ethusini ka Durban ka Phrobentsheng ya KwaZulu-Natal. Ka go šomiša mokgwa wa dinyakišišo wa fenomenokrafiki, sampole ya baithuti ba 15 ba Kreiti ya 9 bao ba ithutelago Dipalo, morutiši o tee wa Dipalo ka go Kreiti ya 9 le morutiši o tee yo e lego hlogo ya lefapha la Dipalo o kgethilwe ka maikemišetšo go tšwa dikolong tše di phagamego tše tharo ka Sedikothutong sa Ethusini. Dikolo tše tharo tše ke sekolo se tee seo se bego se le sa motlolo wa C, se tee sa tlwaelo sa mmušo le se tee ke sekolo sa phraebete. Sehlopha seo se nepišitšwego le dipoledišano tša dipotšišo tšeo di beakantšwego peleng di šomišitšwe go kgoboketša tshedimošo go tšwa go bakgathatema. Tshedimošo e sekasekilwe ka go ngwalolla dipoledišano tša dipotšišo tša sehlopha seo se nepišitšwego le tšeo di beakantšwego peleng; ka morago ga fao, dingwalollo tše di filwe dikhoutu go ya ka merero le magoro a tšona. Dipoelo tša dinyakišišo tše di laetša barutiši le baithuti ba bona gore go se šome gabotse ga baithuti ka thutong ya Dipalo ka go Kreiti ya 9 ke ka lebaka la ditlhohlo tša lenaneothuto la Setatamente sa Melawana ya Tekolo ya Lenaneothuto (CAPS), go swana le diteng tša thuto tšeo di tletšego kudu le tšeo di gatetšego pele kudu, le nako ye e sego ya lekana ya go fetša lenaneothuto. Thekgo ya batswadi, Seisemane (polelo ya gae ya tlaleletšo) bjalo ka polelo ya go ithuta le ya go ruta, mekgwa ya go ruta (ya go swana le mokgwa wa go bolela le go ngwala ka tšhooko letlapeng), le maikutlo ao a sego a loka a baithuti go Dipalo le tšona di filwe bjalo ka dilo tšeo di bakago se. Go dikutollo tše, ditšhišinyo tša go etela leswa lenaneothuto la CAPS le makala a mangwe ao o hlokago šedi, gammogo le dihlogotaba tša dinyakišišo tša ka moso le tšona di ile tša šišinywa. / Lolu cwaningo luphenye imibono yothisha kanye neyabafundi yezimbangela zokungasebenzi kahle esifundweni seziBalo kubafundi beBanga le-9 kuSekhethi yase-Ethusini eThekweni esifundazweni saKwaZulu-Natali. Ngokusebenzisa isakhiwo sokucwaninga esisezingeni eliphezulu, isampula yabafundi abayi -15 beBanga le-9 abathatha iziBalo, uthisha oyedwa weBanga le-9 kanye nomphathi oyodwa womnyango weziBalo bakhethwa ngenhloso ezikoleni ezintathu eziphezulu kuSekhethi yase-Ethusini. Lezi zikole ezintathu esisodwa semodeli C, esisodwa sikahulumeni ojwayelekile kanye nesisodwa isikole esizimele. Iqembu lokugxila kanye nezingxoxo ezihlelwe kahle zisetshenzisiwe ukuqoqa imininingwane evela kubahlanganyeli. Imininingwane yahlaziywa ngokubhalwa kweqembu okugxilwe kulo kanye nezingxoxo ezihlelwe kahle; emva kwalokho, imibhalo ivezwe ngendlela engaqondile ezindikimbeni nasezigabeni. Imiphumela yalolu cwaningo ikhombisa ukuthi othisha kanye nabafundi babona ukuthi ukungasebenzi kahle esifundweni seziBalo kuBanga lesi-9 kungenxa yezinselela yezifundo ezibandakanya isifundo esifundwayo seCAPS, njengokuqukethwe okugcwele kakhulu nokuthuthukile, kanye nesikhathi esinganele sokuphothula izifundo ezibandakanya isifundo esifundwayo. Ukusekelwa kwabazali, isiNgisi (ulimi lokuqala olwengeziwe) njengolimi lokufunda nokufundisa, izindlela zokufundisa (njengeyokufundisa lapho uthisha ebhala ebhodini), kanye nesimo sengqondo esingesihle sabafundi maqondana neziBalo nakho kuveziwe. Ukusuka kulokhu okutholakele, kuphakanyiswe ukuphinda kuphinde kufundwe izifundo ezibandakanya isifundo esifundwayo seCAPS kanye nezinye izindawo ezidingeka, kanye nezihloko zocwaningo lwakusasa ziphakanyisiwe. / Curriculum and Instructional Studies / M. Ed. (Curriculum and Instructional Studies)

The growing diversity of school populations around the world means that for many learners the language of learning in mainstream classrooms is not their first language. The researcher would submit that content-based second language learning in a context such as a Science classroom is considered advantageous as it enables the learner to manipulate a target language such as English in a way which is meaningful. However, Science learners who have yet to achieve communicative competence in English are disadvantaged when it comes to developing a deep understanding of scientific concepts. Many mainstream Science educators have concerns about this significant group of learners who can be left on the periphery of the class to cope as best as they can. Very often educators aim to meet the needs of English Second Language (ESL) learners without any specific knowledge of the strategies which would enhance learning and ensure that learning environments encourage participation and interaction. The learners themselves have not only to deal with language and sociocultural issues but must face the cognitive demands of Science including negotiating its specialized language. The aim of this study was to investigate how the use of English as a language of learning and teaching Science in rural secondary schools in the Vlakfontein Circuit of the Limpopo Capricorn District, influenced the ability of Grade 8 students to learn Science. The focus was on the Grade 8 classes since they are at the threshold of their educational pursuit. The study had two main purposes. The first goal was to describe the current situation with respect to rural secondary school learners and their educators in selected learning environments in Vlakfontein Circuit. The second goal was to bring about improvement in the learners’ situations by employing specifically designed interventions. The study had three focal areas: the language; the teaching and learning environment; and the ESL learner. The investigation was conducted in disadvantaged rural secondary schools in the Vlakfontein Circuit in the Limpopo province. Observations of the Science classes revealed that, even for the learners with very limited English language proficiency there was little ESL specialist support available. The data analyzed was collected using a variety of data collection tools. The main data generation tools were observation, semi-structured interviews and questionnaires. The analysis revealed that learners were lacking in Science register (terminology). This study also found out that non-technical language used in Science lessons affected the learners’ understanding much more than the educators were aware. Educators’ attitudes and beliefs strongly influenced the interaction and participation of ESL learners in Science classrooms. The study also revealed that developing language skills prevented ESL learners from asking questions and answering questions in class and academic progress in Science was impeded by limited opportunities for ESL learners to clarify their understanding. Further, the investigation established that achievement in Science and in education overall was affected by assessment instruments which were infused with specific linguistic or cultural knowledge. The study arrived at a conclusion that the needs of the research students could not be met by a programme based on the traditional format of ESP teaching. Hence, the researcher has recommended a Science Based English Programme (SBEP) which encapsulates several adjustments in orientation methods and materials to meet the ESL learners’ needs. This kind of learning-centred arrangement will allow not only efficiency of SBEP instruction but also allow the kinds of activities that may not be possible in groups with a wide dispersion of interests. However, the extent to which the research learners encountered difficulties with vocabulary suggests that there is a need to investigate more effective methods of dealing with this issue. Needed research could also be directed into the development of a Science glossary with appropriate language levels for ESL learners. This could include technical scientific terms with examples of how terms can be used.From the outcome of the interviews with educators, it is evident that further research is needed concerning the educators’ English proficiency and Science competence in ESL situations. / Teacher Education / D. Ed. (Didactics)