The class of inertial sensors have undergone a revolution in the last decades. From the technological standpoint, this transformation has brought a dramatic reduction in both size and cost after the advent of micromachined electro–mechanical system (MEMS) technology. From an application standpoint, they have spread out from the traditional aeronautical and naval fields (i.e. inertial navigation systems) to a plethora of new and different areas, e.g. automotive, robotics and clinical to name a few. Nowadays, inertial sensors are available on the market as system on chip (SoC) that are small enough to be unobtrusively attached to any system, including the human body, and their presence in consumer products has become a commonplace (e.g. in smartphones) virtually electing them as a top player in the upcoming wearables era. However, miniaturized inertial sensors come with inherent limitations that are mainly found in the reduced performance in terms of noise. Besides, the more recently introduced applications are still far from being mature (e.g. human motion capture). As a consequence, there is a number of open challenges related to the use of this evolved technology and that spans from signal processing and sensor fusion of noisy measurements to the improvement of existing algorithms and expansion to unexplored areas of application. The main objective of this work is to improve the state of the art with respect to some of these challenges. First, the dissertation examines the problem of establishing accuracy in measuring orientation for current inertial sensor fusion algorithms. A methodology based on the use of a robotic manipulator is presented and confidence intervals of static and dynamic performance are established. Then, the thesis discusses the problems of calibration and motion tracking with inertial sensors in two different contexts of application, i.e. ground mobile robotics and biomedical research. Regarding robotics, the setup explored in this thesis consisted of an omnidirectional wheeled platform, equipped with an inertial sensor and wheel encoders, intended to be navigated in a industrial setting. While this class of robots has superior mobility characteristics, difficulties related to their autonomous navigation prevent their widespread use in the research and industry. To overcome this limitation, a number of sensor fusion problems are presented and a solution proposed in order to self calibrate the platform and ameliorate accuracy of the navigation through robustness, e.g. against wheel slippage. Regarding the biomedical application, inertial sensors were used to reconstruct the motion of children (6-7 years old) in a daily life scenario. To this purpose, a novel calibration procedure and algorithms are introduced to improve accuracy over state of the artmethods. The outcome of this research will permit the investigation of motor disorders at an earlier stage of development than is currently possible (e.g. for the case of autism spectrum disorder).

On inertial sensing of motion: validation, calibration and tracking with wearable devices in robotics and biomedical applications / Luca Ricci , 2015 Jun 11. 27. ciclo

On inertial sensing of motion: validation, calibration and tracking with wearable devices in robotics and biomedical applications

2015-06-11

Abstract

The class of inertial sensors have undergone a revolution in the last decades. From the technological standpoint, this transformation has brought a dramatic reduction in both size and cost after the advent of micromachined electro–mechanical system (MEMS) technology. From an application standpoint, they have spread out from the traditional aeronautical and naval fields (i.e. inertial navigation systems) to a plethora of new and different areas, e.g. automotive, robotics and clinical to name a few. Nowadays, inertial sensors are available on the market as system on chip (SoC) that are small enough to be unobtrusively attached to any system, including the human body, and their presence in consumer products has become a commonplace (e.g. in smartphones) virtually electing them as a top player in the upcoming wearables era. However, miniaturized inertial sensors come with inherent limitations that are mainly found in the reduced performance in terms of noise. Besides, the more recently introduced applications are still far from being mature (e.g. human motion capture). As a consequence, there is a number of open challenges related to the use of this evolved technology and that spans from signal processing and sensor fusion of noisy measurements to the improvement of existing algorithms and expansion to unexplored areas of application. The main objective of this work is to improve the state of the art with respect to some of these challenges. First, the dissertation examines the problem of establishing accuracy in measuring orientation for current inertial sensor fusion algorithms. A methodology based on the use of a robotic manipulator is presented and confidence intervals of static and dynamic performance are established. Then, the thesis discusses the problems of calibration and motion tracking with inertial sensors in two different contexts of application, i.e. ground mobile robotics and biomedical research. Regarding robotics, the setup explored in this thesis consisted of an omnidirectional wheeled platform, equipped with an inertial sensor and wheel encoders, intended to be navigated in a industrial setting. While this class of robots has superior mobility characteristics, difficulties related to their autonomous navigation prevent their widespread use in the research and industry. To overcome this limitation, a number of sensor fusion problems are presented and a solution proposed in order to self calibrate the platform and ameliorate accuracy of the navigation through robustness, e.g. against wheel slippage. Regarding the biomedical application, inertial sensors were used to reconstruct the motion of children (6-7 years old) in a daily life scenario. To this purpose, a novel calibration procedure and algorithms are introduced to improve accuracy over state of the artmethods. The outcome of this research will permit the investigation of motor disorders at an earlier stage of development than is currently possible (e.g. for the case of autism spectrum disorder).
11-giu-2015
IMU; inertial tracking; motion capture; children motion; robot navigation
On inertial sensing of motion: validation, calibration and tracking with wearable devices in robotics and biomedical applications / Luca Ricci , 2015 Jun 11. 27. ciclo
File in questo prodotto:
File Dimensione Formato  
DT_115_RicciLuca.pdf

accesso aperto

Tipologia: Tesi di dottorato
Licenza: Creative commons
Dimensione 7.74 MB
Formato Adobe PDF
7.74 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12610/68767
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact