Biomedical Engineering Reference
In-Depth Information
As for the joint torques and forces, muscle tensions are good indicators of gait
efficiency and can thus be used to determine subtle changes in gait patterns that
kinematic data fail to identify. It could thus be used to evaluate if the user's gait is as
efficient in VR compared to natural walking.
8.7 Conclusion
J. E. Marey and E. Muybridge were the first researchers who proposed objective
measurements of animal and human locomotion [
20
]. Nowadays, numerous systems
exist to analyse and measure human gait. The biomechanics community is still very
active in this domain and collaboration with other domains will certainly lead to new
systems, such as using depth-cameras (Kinect of Microsoft) or inertial sensors. It is
difficult to predict what will be the future in this domain, but many researchers tend to
propose non-invasive and light systems associated with more and more sophisticated
numerical models in order to access to new parameters. Musculoskeletal models are
clearly a step forward in this domain but many researches have to be carried-out in this
domain for calibration and validation. The Table below summarizes the parameters
introduced in this chapter and their potential use in walking in VR (Table
8.1
):
•
sensing the user's motion,
•
delivering the most appropriate and accurate multisensory feedbacks,
•
and evaluating the naturalness of the interaction for the user compared to reference
values (as those reported in Chapter “Biomechanics of walking in real world”).
Systems deliver more and more data (global parameters, joint positions, angles,
torques and muscle forces) and a new problem occur: how to deal with all these
parameters? In virtual reality researchers try to provide the user with more realistic
feedbacks which could rely on using such big amount of knowledge. Getting more
accurate data on the user's gait could help to select and compute the corresponding
feedbacks. Let us consider the walk-in-place approach currently used in VR [
36
]. In
this approach, signal processing is applied to the orientation of the head depending
time. Extending this approach to more complex parameters would enable to deal
with more complex situations and behaviours, such as getting up and down stairs,
avoiding obstacles or taking specific gait style into account. As for future steps, one
main challenge therefore remains to be able to compile all the available data and to
compute the most appropriate feedbacks in real-time.
References
1. Amarantini D, Rao G, Berton E (2010) A two-step EMG-and-optimization process to estimate
muscle force during dynamic movement. J Biomech 43(9):1827-1830
2. Auvinet B, Berrut G, Touzard C, Moutel L, Collet N, Chaleil D, Barrey E (2002) Reference
data for normal subjects obtained with an accelerometric device. Gait Posture 16:124-134
