20.500.12556/RUL-171477 Napoved trajektorije šestosnega robota na osnovi merjenega pospeška Trajectory prediction of a six-axis robot based on measured acceleration Obravnavali smo kinematiko šestosnega robota in raziskali možnost neodvisne napovedi njegove trajektorije gibanja zgolj na podlagi merjenih pospeškov v treh ortogonalnih smereh. Cilj naloge je bil razviti metodologijo, ki bo omogočala rekonstrukcijo gibanja robota brez neposrednega dostopa do podatkov krmilnika, ki smo jih uporabili za izračun referenčne trajektorije z direktno kinematiko. Kalibracijo pospeškomera smo izvedli v širokem frekvenčnem območju, s čimer smo zagotovili natančnost zajema signalov in prilagodili občutljivost senzorja. Pospeške smo zajemali med različnimi gibi robota, pridobljene signale pa analizirali v frekvenčni domeni s pomočjo Fourierove transformacije. Poseben poudarek je namenjen filtriranju šuma in izbiri optimalnih numeričnih filtrov in metod. Rezultate obdelave smo uporabili za napoved trajektorije, ki smo jo primerjali s podatki iz krmilnika robota. S tem smo ocenili natančnost in uporabnost predlagane metode ter identificirali morebitne omejitve. We analyzed the kinematics of a six-axis robot and explored the possibility of independently predicting its motion trajectory solely based on measured accelerations in three orthogonal directions. The aim of the thesis was to develop a methodology that enables the reconstruction of the robot's motion without direct access to controller data, which we used to calculate the reference trajectory through direct kinematics. The accelerometer was calibrated across a wide frequency range, ensuring accurate signal acquisition and adjusting the sensor's sensitivity. Accelerations were recorded during various robot movements, and the acquired signals were analyzed in the frequency domain using Fourier transform. Special attention was given to noise filtering and the selection of optimal numerical filters and methods. The processed data was used to predict the trajectory, which was compared to the data obtained from the robot's controller. This allowed us to evaluate the accuracy and applicability of the proposed method and identify potential limitations. kinematika MEMS inercijska zaznavala kapacitivnostni pospeškomeri frekvenčna analiza numerične metode numerično filtriranje kinematics MEMS inertial sensors capacitive accelerometers frequency analysis numerical methods numerical filtering true false false Slovenski jezik Angleški jezik Diplomsko delo/naloga 2025-08-27 08:31:27 2025-08-27 08:31:29 2025-09-08 09:44:22 0000-00-00 00:00:00 2025 0 0 X, 26 f. 2025 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 007.52:531.113:517.443(043.2) 274782 247991299 43217.pdf 43217.pdf 1 63EB77CB14FE38B30A17D4EA65E2AB73 242fb1e3bfece33a903e17095e7b82351b72b2d3c515ca2fb10559828934654a 568ef1c8-830f-11f0-9328-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=214808 Fakulteta za strojništvo 0 0 0