Kinematic Reconstruction and Occupational Biomechanics

This research line deals with the use of Inertial Units for the reconstruction of the human body kinematic pose and the addition of sensor fusion techniques for improving the estimation thanks to different sensors. These techniques are aimed at Occupational Biomechanics analysis.


We have also presented these result at the national workshop ERGANE 2015, and at the IEEE MISS Summer School in September 2016.


The objective is the use of Bayesian filtering for the reconstruction of the pose from multiple, possibly unsynchronized IMU, taking advantage of biomechanical and task constraint. These techniques have been investigated for the purpose of Occupational Biomechanics. Sensor fusion can be then used for improving the estimate or resetting it at lower rate.


Modeling via Denavit-Hartenberg chains, using probabilistic graphical models (PGM) with non linear modeling via Unscented Kalman Filtering. Lab assessment comparing Marker based tracking with inertial measures done with custom wireless boards.


The following results have been achieved:

  1. Use of PGM for expressing the kinematic uncertainty (ICRA 2014)
  2. Fusion with ground sensors with kinematic closure for improving estimate in Rowing (JST 2015)
  3. Modelling and estimation of upper arm with the Clavicle-joint (SISY 2013)
  4. Practical use of the system for Occupational Biomechanics in comparison to RULA (International Journal of Industrial Ergonomics 2015)
  5. Use in the domain of Human-Robot interaction paired with haptic suit (ISR 2016)
  6. Integration with a haptic interface for rehabilitation robotics (MFI 2016)



Italian Brochure of the SMOOTI system




For the moment we have not released the Source Code but there are some useful bits:

  • Tool for comparison of non linear functions and Multivariate Normals (Matlab) (github)

Funding Projects

This activity has been funded by the Italian Ministry of Health CCM project ERGANE for biomechanics of workers, and by Telecom Italia in the SMOOTI project. This research is continuing in the ECHORD++ project MOTORE++ for combining inertial measurements with the MOTORE planar rehabilitation haptic interface.

We also acknowledge the IEEE RAS Technical Committee on Human Movement Understanding for organizing the ICRA Workshops.

In the News

  • Launch of the INAIL BRIC 24 project (July 2017): news
  • Radio24 about Smooti (September 2016) : link
  • End of the Smooti Telecom project (July 2016): link

Journal Publications

  • Survey of motion tracking methods based on inertial sensors: a focus on upper limb human motion
    Filippeschi Alessandro, Schmitz Norbert, Miezal Markus, Bleser Gabriele, Ruffaldi Emanuele & Stricker Didier
    Sensors (in press) 1424-8220, 1 ,2017,doi:10.3390/s17061257 Code (github) Dataset (Zenodo)
  • Ruffaldi E., Peppoloni L. & Filippeschi A. (2015). Sensor fusion for complex articulated body tracking applied in rowing. Journal of Sport Engineering and Technology, 229(2), (pp. 92–102). doi:10.1177/1754337115583199
  • Peppoloni L., Filippeschi A., Ruffaldi E. & Avizzano C.A. (2015). A novel wearable system for the online assessment of risk for biomechanical load in repetitive efforts. International Journal of Industrial Ergonomics, (in press), . doi:10.1016/j.ergon.2015.07.002

Conference Publications

  • Saracino L.A., Ruffaldi E., Graziano A. & Avizzano C.A. (2016). Fusion of wearable sensors and mobile haptic robot for the assessment in upper limb rehabilitation. In IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI)
  • Graziano A., Tripicchio P., Ruffaldi E. & Avizzano C.A. (2016). A wireless haptic data suit for controlling humanoid robots. In 47th International Symposium on Robotics (ISR). IEEE.
  • Peppoloni L., Filippeschi A. & Ruffaldi E. (2014). Assessment of task ergonomics with an upper limb wearable device. In Control and Automation (MED), 2014 22nd Mediterranean Conference of (pp. 340-345). . doi:10.1109/MED.2014.6961394
  • Avizzano C.A., Ruffaldi E. & Bergamasco M. (2014). A novel wearable biometric capture system. In Control and Automation (MED), 2014 22nd Mediterranean Conference of(pp. 351-355). . doi:10.1109/MED.2014.6961396
  • Ruffaldi E., Peppoloni L., Filippeschi A. & Avizzano C.A. (2014). A novel approach to motion tracking with wearable sensors based on Probabilistic Graphical Models. InRobotics and Automation (ICRA), 2014 IEEE International Conference on (pp. 1247-1252). . doi:10.1109/ICRA.2014.6907013
  • Peppoloni L., Filippeschi A., Ruffaldi E. & Avizzano C.A. (2013). A novel 7 degrees of freedom model for upper limb kinematic reconstruction based on wearable sensors. InIntelligent Systems and Informatics (SISY), 2013 IEEE 11th International Symposium on (pp. 105-110). . doi:10.1109/SISY.2013.6662551

Workshop Posters

  • Ruffaldi E. (2015). Reconstruction and Analysis in the ERGANE system, ERGANE Workshop, Pisa, Italy
  • Peppoloni L., Ruffaldi E., Filippeschi A. & Avizzano C.A. (2015). Wearable solution for online assessment of biomechanical load risks.At Bridging Gaps between Computational Biomechanics and Robotics: Theory, Tools, and Applications, Tutorial inside IEEE RAS ICRA. Seattle,Washington
  • Peppoloni L., Filippeschi A. & Ruffaldi E. (2014). Sensor fusion for complex articulated body tracking applied in rowing.In ICRA Workshop on Latest Advances on Natural Motion Understanding and Human Motion Synthesis
  • Peppoloni L., Filippeschi A. & Ruffaldi E. (2013). Motion Tracking for portable biomechanic measures.In ICRA Workshop on Computational Techniques in Natural Motion Analysis and Reconstruction (Slides)

We acknowledge the IEEE RAS Technical Committee on Human Movement Understanding for organizing the ICRA Workshops.

Link to original PERCRO page:

Related Project by the Scientific Community

The following video is an example of commercial solutions in the same topic area (note that we have not been involved in the development of such product):