The Experts below are selected from a list of 3738 Experts worldwide ranked by ideXlab platform
Tachtsis Eleftherios - One of the best experts on this subject based on the ideXlab platform.
-
On special partitions of Dedekind- and Russell-Sets
'World Scientific Pub Co Pte Lt', 2012Co-Authors: Herrlich Horst, Howard Paul, Tachtsis EleftheriosAbstract:summary:A Russell Set is a Set which can be written as the union of a countable Pairwise Disjoint Set of pairs no infinite subSet of which has a choice function and a Russell cardinal is the cardinal number of a Russell Set. We show that if a Russell cardinal $a$ has a ternary partition (see Section 1, Definition 2) then the Russell cardinal $a+2$ fails to have such a partition. In fact, we prove that if a ZF-model contains a Russell Set, then it contains Russell Sets with ternary partitions as well as Russell Sets without ternary partitions. We then consider generalizations of this result
Herrlich Horst - One of the best experts on this subject based on the ideXlab platform.
-
On special partitions of Dedekind- and Russell-Sets
'World Scientific Pub Co Pte Lt', 2012Co-Authors: Herrlich Horst, Howard Paul, Tachtsis EleftheriosAbstract:summary:A Russell Set is a Set which can be written as the union of a countable Pairwise Disjoint Set of pairs no infinite subSet of which has a choice function and a Russell cardinal is the cardinal number of a Russell Set. We show that if a Russell cardinal $a$ has a ternary partition (see Section 1, Definition 2) then the Russell cardinal $a+2$ fails to have such a partition. In fact, we prove that if a ZF-model contains a Russell Set, then it contains Russell Sets with ternary partitions as well as Russell Sets without ternary partitions. We then consider generalizations of this result
Howard Paul - One of the best experts on this subject based on the ideXlab platform.
-
On special partitions of Dedekind- and Russell-Sets
'World Scientific Pub Co Pte Lt', 2012Co-Authors: Herrlich Horst, Howard Paul, Tachtsis EleftheriosAbstract:summary:A Russell Set is a Set which can be written as the union of a countable Pairwise Disjoint Set of pairs no infinite subSet of which has a choice function and a Russell cardinal is the cardinal number of a Russell Set. We show that if a Russell cardinal $a$ has a ternary partition (see Section 1, Definition 2) then the Russell cardinal $a+2$ fails to have such a partition. In fact, we prove that if a ZF-model contains a Russell Set, then it contains Russell Sets with ternary partitions as well as Russell Sets without ternary partitions. We then consider generalizations of this result
Koložvari Andrej - One of the best experts on this subject based on the ideXlab platform.
-
Wheelchair design as a cyber-physical system and analysis of its impact on the organization of the rehabilitation process
2021Co-Authors: Koložvari AndrejAbstract:V tem delu je opredeljena možnost razvoja invalidskega vozička kot kiberfizičnega sistema, ki je v tesni povezavi z internetom in njegovimi uporabniki. Definirana je struktura kiberfizičnega sistema s ključnimi povratnimi zankami, ki omogočajo upravljanje vozička s pomočjo govora. Preučena je možnost uporabe v rehabilitacijskih procesih ob višji stopnji avtonomnosti bolnikov. Izvedena je bila analiza ročno upravljanega invalidskega vozička kot izhodišče za razvoj prototipov govorno upravljanega vozička. Opredeljen je bil prototipni pristop k razvoju kiberfizičnega vozička, ki je potekal v več fazah. Razvitih je bilo več manjših prototipov in dva prototipa realne velikosti. Opisan je razvoj manjših prototipov kakor tudi prenos funkcionalnosti na dva prototipa realne velikosti. Podana je zasnova govorno upravljanega kiberfizičnega sistema invalidskega vozička za ljudi s posebnimi potrebami. Razviti sta bili dve rešitvi v obliki prototipa realne velikosti. Prvi prototip realne velikosti za razpoznavo govora izkorišča le en oblačni informacijski sistem za razpoznavo govora, hkrati pa vključuje le osnovna tipala za pomoč pri navigaciji. Drugi prototip realne velikosti ima vključena dva oblačna informacijska sistema za razpoznavo govora ter več tipal, ki zagotavljajo določeno stopnjo avtonomnosti vozička. Opisan je programski sveženj, ki je omogočil realizacijo sistema in prenos sporočil med klientom, strežnikom in oblačnim informacijskim sistemom. Prav tako je podrobno opisan sveženj strojne opreme, ki omogoča hiter prototipni razvoj in enostavno povezljivost z informacijskim oblakom. Uporabniški vmesnik sistema omogoča govorni vhod in nadzor s pomočjo spletne kamere. Sistem je zasnovan tako, da je omogočeno oddaljeno posodabljanje programske opreme. Na prototipih realne velikosti je bila izvedena analiza natančnosti izvedbe izrečenih ukazov s pomočjo testnih subjektov v laboratorijskem in kliničnem okolju. Prav tako je bila izvedena meritev zakasnitev, kjer se je izkazalo, da je opredeljeni koncept primeren za govorno upravljanje v realnem času. Z namenom večje natančnosti razpoznave govora je bil razvit in uporabljen princip vzporedne oblačne žetve informacijskih sistemov. Za izboljšanje govornega upravljanja je bil definiran vnaprej določen osnovni nabor ukazov. Opredeljeno je oblikovanje tabele zamenjav končnih razpoznanih besed in vmesnih rezultatov razpoznave. Definiran je pogoj parne disjunktnosti elementov množic, ki opredelijo izvedbo posameznega ukaza, ter algoritem združevanja rezultatov več oblačnih sistemov za razpoznavanje govora. Opredeljeni sta izvedljivost uporabe algoritmov za žetev govorne razpoznave v informacijskem oblaku ter teoretična meja natančnosti razpoznave ob podanih pogojih. Uporabnost razvitega kiberfizičnega vozička z možnostjo govornega upravljanja je bila preizkušena v kliničnem okolju. Opredeljen je vpliv na rehabilitacijske procese ob povečanju avtonomije bolnika ter možnosti oddaljenega spremljanja fizioloških parametrov. Z namenom primerjave rezultatov testiranja različnih uporabnikov je bil oblikovan testni poligon za preizkušanje funkcionalnosti. Ob izvedenih preizkusih so bile ugotovljene statistično pomembne razlike v natančnosti izvedbe govornih ukazov, ki so potrdile učinkovitost razvitega algoritma združevanja razpoznav in teoretične predpostavke o možnosti povečanja stopnje natančnosti pri izvedbi govornih ukazov. V sklepnem delu so podana izhodišča za nadaljnje delo predvsem v smislu združevanja več oblačnih informacijskih sistemov za razpoznavo govora in rešitev s področja robnega računalništva.The development of a cyber-physical wheelchair for disabled persons that is tightly interconnected with the internet and its users is considered in the present thesis. The cyber-physical system structure is defined with key feedback loops, which enable speech-controlled navigation of the wheelchair. Application within rehabilitation processes is considered since the developed system provides for the higher autonomy of patients. An analysis of a hand-steered wheelchair was performed as the starting point for the prototype development. The prototype development of the cyber-physical wheelchair was described. Several smaller prototypes were developed. In the second phase, two full-sized prototypes were made. The development of small prototypes is described as well as the transfer of functionality from small to two full-sized prototypes. The design of the speech-controlled cyber-physical wheelchair system for disabled persons is described. Two full-sized solutions were developed. The first full-sized prototype uses only one cloud information system for speech recognition, incorporating only a basic Set of sensors to aid in the navigation task. The second such prototype uses two different cloud information systems for speech recognition and includes more sensors, which provide a certain level of wheelchair autonomy. The software stack used is described as well as the process of message interchange between client, server, and cloud information system. The hardware stack is also describedit enables rapid prototype development and easy interconnection with the information cloud. The user interface of the system enables spoken input as well as control by a web camera. The system is designed in such a way that remote software updates are possible. With the full-sized prototypes, an analysis of the accuracy of the commands issued by test subjects was performed in the laboratory and clinical Setup. Latency measurements were performed in which the values proved that the concept is appropriate for the speech control of navigation in real time. To improve the accuracy in the interpretation of issued commands, the principle of parallel cloud harvesting was introduced. A predefined Set of commands was determined in order to improve speech recognition and command interpretation. Synonym tables were defined for the final and intermediate results of speech recognition. A condition of Pairwise Disjoint Set elements was introduced, which determine the execution of specific commands as well as the merging algorithm, which employs the results of several cloud-based information systems for speech recognition. The applicability of the algorithms for speech recognition cloud harvesting was analysed, and the theoretical boundary of system accuracy was defined. The usefulness of the speech-controlled cyber-physical wheelchair was tested in a clinical environment. The impact on the rehabilitation processes with the increased patient autonomy was analysed as well as remote monitoring of biomedical parameters. A test route was formed to compare the results of users’ testing and functionality. With the experiments, the statistically significant differences in the accuracy of command interpretation were indicated, which confirmed the effectiveness of the developed cloud-harvesting algorithm which merged the results of cloud speech recognition. Furthermore, the theoretical consideration of the accuracy improvement was confirmed. In the final part of the dissertation, the directions for future research are outlined, which consider the merging of more cloud-based information systems for speech recognition as well as the solutions from the field of cloud computing
