Sports Innovation, Technology And Research

Dominic F L Southgate, Peter R N Childs;Anthony M J Bull

• Verlag: WSPC (EUROPE)
• Jahr: 2016
• ISBN:
  9781786340436
• Thema: Tecnología e ingeniería
• Thema: Ciencias biomédicas

Sports Innovation, Technology and Research gives an insight into recent research and design projects at Imperial College London. It presents the on-going development of a diverse range of areas from elite rowing performance to impact protection

Chapter 1

Motion Analysis

in Sport

Dr. Dominic F L Southgate, Dr. Joe A I Prinold, and Dr. Robert A Weinert-Aplin

Imperial College London

Introduction

It is hard to overstate the contribution that a skilled coach can make to an athlete’s performance and experience counts for a great deal in the realm of competitive sport. That said, coaches can still have questions when it comes to optimal performance and technique — sometimes they may have a ‘feel’ for what is appropriate but not have any data to back it up and sometimes there are complete unknowns. These questions may be on the topic of absolute competitive performance (e.g. running faster, jumping higher, etc.), injury prevention (during training or competition), equipment interaction (e.g. optimal bat/club/racket swing) or team interaction (e.g. rowing synchronicity).

Luckily, over the past couple of decades, various tools have been developed to aid in the analysis of human motion which can help to answer some of these problems. Generally, these tools are used to track the motion of ‘segments’ of the body, so that joint angle/power/contact force etc. can be calculated. The answers sought may come directly from the technology/device without the need for further data processing (although processing will be required at some point within the device) or, depending on the application, more complex musculoskeletal modelling may be required. This type of modelling is a vast research topic in itself, requiring an even more diverse range of inputs.

There have been a number of fields driving these technologies, including medical/clinical applications and the computer game and film making industries, as well as some sports themselves at the elite or professional level. Indeed, between and within each field different technologies have developed, each with their own advantages and disadvantages. A brief description of some common types is given in the following section.

Motion Capture Technology

Inertial Measurement Units consist of a number of micromechanical devices which in their most basic form will be just accelerometers but then also include gyroscopes and magnetometers as they become more sophisticated. Accelerometers can be used to measure angular rotations and work well on a static subject but for dynamic motions, the outputs become more complex and harder to interpret. The addition of gyroscopes and magnetometers can overcome this when combined with sensor fusion algorithms, and this is what is now commonly found in most smartphones. There are commercial systems available specifically for motion capture of the body (such as XSens), as well as many generic units that can be applied to the body. The main advantages of these devices are cost and simplicity; however drawbacks can include difficulties in synchronising multiple devices and lack of a simple output of spatial location.

Electromagnetic Systems use a small sensing coil within a well-defined electromagnetic field to determine position and orientation. These have been much used in motion analysis applications and commercial systems do exist for this purpose (such as those made by Polhemus and Ascension Technologies). Advantages of these systems include being able to output location and orientation from a single sensor and the ability to easily synchronise multiple sensors. However, a major disadvantage is that any metal within the magnetic field can introduce errors into the measurements, and the systems are normally limited to a small number of sensors (and therefore body segments that can be measured).

Markerless Camera Techniques are gaining in popularity and the simplest form is 2D capture from a single camera, commonly used in Video Vector Analysis. However, more sophisticated forms are emerging that reconstruct 3D motion from regular video images, based on sophisticated image processing. A variation is the Microsoft Kinect system which uses a pattern of infrared dots combined with two cameras to give a quasi-3D motion capture, and which is finding many applications due to its availability and adaptability. The main drawback with this technology is the level of accuracy, as with many applications the aim is track the motion of the bone in the segment as it is the rigid body. Soft tissue around a limb segment can move and change shape, thus creating errors for the system in determining the position of the bone.

Medical Imaging includes a number of different technologies, such as MRI, CT, ultrasound, and fluoroscopy and each has its own use for visualising specific structures within the body. This makes it ideal for visualising the exact movement of soft tissue relative to bone, for example, and in this sense it can be considered the gold standard. However, its use in sport is generally limited due to the highly constrained environments in which they must be used. Despite this, it does see use in certain applications and at Imperial open MRI has been used to investigate the motion of the spine in rowers during the different phases of the stroke.

Optical Motion Capture is used here to refer to systems that track either active or passive markers placed on the body and will be the main focus of this chapter due to the experience with this type of system at Imperial. These types of system generally have a series of cameras all focused on the ‘capture volume’ and each one records the position of the markers in 2D as single points moving in space. Through an initial calibration process, the system knows the position of each camera and is able to piece together all of the 2D images to locate each of the markers in 3D space. As each marker is seen as a single point in space, three markers are needed, on each segment to completely define its position and orientation in the capture volume, however, it is usually possible to have a very large number of markers.

This type of system is generally very expensive but the accuracy of marker positions and flexibility of the set up has made this the gold standard for motion capture. There are still issues though; the markers are normally adhered to the skin with tape or glue and skin motion artefacts can be a problem when there is a lot of soft tissue overlying the bony landmarks. Experiments have been conducted (by other groups) using transcutaneous bone pins to overcome this problem but this is obviously highly invasive and unlikely to be used in a sports context.

Uptake of this technology for use in sports has not been universal and reasons for this can include the value of the outputs relative to the effort and the costs required to obtain motion analysis systems and to become trained in their use and interpreting the results. Many systems still require a very high level of engineering or scientific expertise and sports teams or national bodies often collaborate with academic institutions for this reason.

There is also the perennial issue of trying to capture motion in a realistic scenario. Many optical motion capture systems are highly affected by sunlight and so attempting to record outside can be problematic. Water, snow and ice (as well as any shiny object) can cause reflections and so these types of environments are generally precluded. Other elements of the system, such as force-plates, may also be non-portable and so it is often in the motion analysis laboratory where experiments are performed. This can be very different to an athlete’s training environment (and competition even more so) and this may limit the value of the data obtained. Despite this, there is still a wealth of useful information that can be gained from motion capture experiments, some examples of which will be detailed throughout this chapter.

Kinematics

One of the most important outputs from motion analysis in sport is kinematics — i.e. the angles of the joints and how they are changing with time throughout a motion. At a superficial level, this may seem straightforward, and in fact a simple analysis can be achieved using a device such as a goniometer (essentially a protractor placed over the joint). This can potentially give useful measures, where it can be assumed there is a single axis and it is perpendicular to the plane of joint motion, such as at the elbow or knee. However, it quickly becomes redundant for fast activities and where there is highly 3D motion, such as at the shoulder joint.

To measure 3D motion of a joint, it becomes a necessity to define a co-ordinate system for the body segment on either side of the joint. That is, an ‘x, y, z’ axis system that is fixed to points on each segment, such that when they move, the angles between the respective axes can be measured. Generating these co-ordinate systems relies on specific landmarks which are normally chosen such as to minimise the motion of the skin (th...

Inhaltsverzeichnis

• Cover
• Halftitle
• Title
• Copyright
• Contents
• About the Editors
• Introduction
• Preface: Innovation in Rio Tinto Chris Goodes
• Part A — Innovation and Sports Research
• Part B — RTSIC Projects 2012–2013
• Index of Poems
• Index of Contributors
• Index of Subjects