Gabriele Bleser

Innovative system for real-time ergonomic feedback in industrial manufacturing.

This work presents a system that permits a real-time ergonomic assessment of manual tasks in an industrial environment. First, a biomechanical model of the upper body has been developed by using inertial sensors placed at different locations on the upper body. Based on this model, a computerized RULA ergonomic assessment was implemented to permit a global risk assessment of musculoskeletal disorders in real-time. Furthermore, local scores were calculated per segment, e.g. the neck region, and gave information on the local risks for musculoskeletal disorders. Visual information was fed back to the user by using a see-through head mounted display. Additional visual highlighting and auditory warnings were provided when some predefined thresholds were exceeded. In a user study (Nxa0=xa012 participants) a group with the RULA feedback was compared to a control group. Results demonstrate that the real-time ergonomic feedback significantly decreased the outcome of both globally as well as locally hazardous RULA values that are associated with increased risk for musculoskeletal disorders. Task execution time did not differ between groups. The real-time ergonomic tool introduced in this study has the potential to considerably reduce the risk of musculoskeletal disorders in industrial settings. Implications for ergonomics in manufacturing and user feedback modalities are further discussed.

Using egocentric vision to achieve robust inertial body tracking under magnetic disturbances

In the context of a smart user assistance system for industrial manipulation tasks it is necessary to capture motions of the upper body and limbs of the worker in order to derive his or her interactions with the task space. While such capturing technology already exists, the novelty of the proposed work results from the strong requirements of the application context: The method should be flexible and use only on-body sensors, work accurately in industrial environments that suffer from severe magnetic disturbances, and enable consistent registration between the user body frame and the task space. Currently available systems cannot provide this. This paper suggests a novel egocentric solution for visual-inertial upper-body motion tracking based on recursive filtering and model-based sensor fusion. Visual detections of the wrists in the images of a chest-mounted camera are used as substitute for the commonly used magnetometer measurements. The on-body sensor network, the motion capturing system, and the required calibration procedure are described and successful operation is shown in a real industrial environment.

A personalized exercise trainer for the elderly

Physical activity provides many physiological benefits. On the one hand it reduces the risk of disease outcomes. On the other hand it is the basis for proper rehabilitation in case of or after a severe disease. Both aspects are especially important for the elderly population. Within this context, the present paper proposes a personalized, home-based exercise trainer for elderly people. The system is based on a wearable sensor network that enables capturing the users motions. These are then evaluated by comparing them to a prescribed exercise, taking both exercise load and technique into account. Moreover, the results are translated into appropriate feedback to the user to assist the correct exercise execution. A novel part of the system is the generic personalization by means of a supervised teach-in phase.

Activity recognition using biomechanical model based pose estimation

In this paper, a novel activity recognition method based on signal-oriented and model-based features is presented. The model-based features are calculated from shoulder and elbow joint angles and torso orientation, provided by upper-body pose estimation based on a biomechanical body model. The recognition performance of signal-oriented and model-based features is compared within this paper, and the potential of improving recognition accuracy by combining the two approaches is proved: the accuracy increased by 4-6% for certain activities when adding model-based features to the signal-oriented classifier. The presented activity recognition techniques are used for recognizing 9 everyday and fitness activities, and thus can be applied for e.g., fitness applications or in vivo monitoring of patients.

Personalized Physical Activity Monitoring Using Wearable Sensors

It is a well-known fact that exercising helps people improve their overall well-being; both physiological and psychological health. Regular moderate physical activity improves the risk of disease progression, improves the chances for successful rehabilitation, and lowers the levels of stress hormones. Physical fitness can be categorized in cardiovascular fitness, and muscular strength and endurance. A proper balance between aerobic activities and strength exercises are important to maximize the positive effects. This balance is not always easily obtained, so assistance tools are important. Hence, ambient assisted living (AAL) systems that support and motivate balanced training are desirable. This chapter presents methods to provide this, focusing on the methodologies and concepts implemented by the authors in the physical activity monitoring for aging people (PAMAP) platform. The chapter sets the stage for an architecture to provide personalized activity monitoring using a network of wearable sensors, mainly inertial measurement units (IMU). The main focus is then to describe how to do this in a personalizable way: (1) monitoring to provide an estimate of aerobic activities performed, for which a boosting based method to determine activity type, intensity, frequency, and duration is given; (2) supervise and coach strength activities. Here, methodologies are described for obtaining the parameters needed to provide real-time useful feedback to the user about how to exercise safely using the right technique.

Using optical flow as lightweight SLAM alternative

Visual simultaneous localisation and mapping (SLAM) is since the last decades an often addressed problem. Online mapping enables tracking in unknown environments. However, it also suffers from high computational complexity and potential drift. Moreover, in augmented reality applications the map itself is often not needed and the target environment is partially known, e.g. in a few 3D anchor or marker points. In this paper, rather than using SLAM, measurements based on optical flow are introduced. With these measurements, a modified visual-inertial tracking method is derived, which in Monte Carlo simulations reduces the need for 3D points and allows tracking for extended periods of time without any 3D point registrations.

Occlusion-aware video registration for highly non-rigid objects

This paper addresses the problem of video registration for dense non-rigid structure from motion under suboptimal conditions, such as noise, self-occlusions, considerable external occlusions or specularities, i.e. the computation of optical flow between the reference image and each of the subsequent images in a video sequence when the camera observes a highly deformable object. We tackle this challenging task by improving previously proposed variational optimization techniques for multi-frame optical flow (MFOF) through detection, tracking and handling of uncertain flow field estimates. This is based on a novel Bayesian inference approach incorporated into the MFOF. At the same time, computational costs are significantly reduced through iterative pre-computation of the flow fields. As shown through experiments, the resulting method performs superior to other state-of-the-art (MF)OF methods on video sequences showing a highly non-rigidly deforming object with considerable occlusions.

Unsupervised model generation for motion monitoring

This paper addresses two fundamental requirements of full body motion monitoring: (a) the ability to sense the input of the user and (b) the means to interpret the captured input. Appropriate technology in both areas is required for an interactive virtual reality system to provide feedback in a useful and natural way. This paper combines technologies for both areas: It develops a sensor fusion approach for capturing user input based on miniature on-body inertial and magnetic motion sensors. Furthermore, it presents work in progress to automatically generate models for motion patterns from the captured input. The technology is then used and evaluated in the context of a personalized virtual rehabilitation trainer application.