Cloud-based just-in-time computation of  dynamic transformations of biomechanical models

Mykhailov, Dmytro

Домівка
→
Students Research & Project Works | Роботи студентів
→
Факультет прикладних наук
→
Освітня програма наук про дані
→
2024
→
Перегляд матеріалів

dc.contributor.author	Mykhailov, Dmytro
dc.date.accessioned	2024-08-23T08:47:50Z
dc.date.available	2024-08-23T08:47:50Z
dc.date.issued	2024
dc.identifier.citation	Mykhailov Dmytro. Cloud-based just-in-time computation of dynamic transformations of biomechanical models. Ukrainian Catholic University, Faculty of Applied Sciences, Department of Computer Sciences. Lviv 2024, 39 p.	uk
dc.identifier.uri	https://er.ucu.edu.ua/handle/1/4671
dc.language.iso	en	uk
dc.subject	biomechanical models	uk
dc.subject	dynamic transformations of biomechanical models	uk
dc.title	Cloud-based just-in-time computation of dynamic transformations of biomechanical models	uk
dc.type	Preprint	uk
dc.status	Публікується вперше	uk
dc.description.abstracten	To manage the movement control of the human body or robotic systems, the problem of calculating dynamic transformations must be solved. Solving the dynamic trans- formation problem in real-time can be a complex task due to the complex human structure with a large number of degrees of freedom. To solve such problems, a large amount of computing resources or the use of artificial neural networks, which must be created for each model separately, are required. The following study presents a cloud-based solution for the problem of real-time dynamics transformation with the ability to collect the processing data. Such a solution will help solve the problem of limited computing and energy resources on edge devices, as well as systematize and collect data for subsequent research. The proposed solution faces the task of computing dynamic transformation in real time with the conditions of network de- lays. This master thesis proposes to extrapolate previously obtained data by dead reckoning the simulated model using Unscented Kalman Filter.	uk