Kinematics & Placement
Table of Contents
Step 1. Clarify The Problem
When one is attempting to develop a motion tracking system, it is very important to consider the kinematic motion of an human. A human arm, for example, provides nine degrees of motion between the shoulder, elbow, and wrist. One must understand which range of motion they want to measure, and then identify the best way to specifically measure that range of motion without interference from unwanted inputs. In addition, one must fully realize what boundaries of motion are physically capable, and ensure that their sensor placement allows for full ROM measurement.
The following are requirements for sensor placement:
- Sensor placement does not hinder normal human motion
- Sensor placement gives accurate representation of motion desired
- Sensor readings are not affected by external factors (skin, bumps, etc)
- Sensor placement allows for complete range of motion measurements (is not limited to 30 degrees out of a possible 50 degrees, for example).
In this case, it is also necessary to consider several parts of the human body when researching placement and kinematics, with joints of interest including:
- Spine: Lumbar region: Sacrum-L1/L2-L3/L4/L5
- Arm: Shoulder-Elbow-Wrist
- Leg: Hip-Knee-Ankle
Step 2. Search Externally
1) From a PDF by Van Herp, Rowe, Salter, and Paul utilizing the Polhemus 3Space System (Available on EDGE): information on Lumbar/Spine Kinematics
- Title: Three-dimensional lumbar spinal kinematics:
a study of range of movement in 100 healthy subjects aged 20 to 60+ years.
- Bony landmarks and placement: procedure identified by Burton
- Lordosis (curvature of spine) prevents spine and
source from pointing straight at eachother
- Alignment needed to orient axes of source & sensor parallel to anatomical planes of body
- This team utilized adjustable plastic wedges and double sided tape to align:
- Max Joint Angulation Reports (use for benchmarking)
- Discrepancy in values:
- Axial rotation: Russell attributed to skin movement between sensor and underlying vertebrae. More studies are needed
- Overall: Many studies exhibited poor attachment method!
2) From a PDF: Rosen, Perry, Manning, Burns, Hannaford- available on EDGE. Team used a Vicon MT System to capture data, CAD SW to model, and Dynamics SW to derive Euler equations of motion.
- Title: The Human Arm Kinematics and Dynamics
During Daily Activities Toward a 7 DOF Upper Limb Powered Exoskeleton
- Motions of Interest:
- Elbow flexion/extension
- Elbow rotation (supination/pronation)
- Shoulder adduction/abduction
- Shoulder flexion/extension
- Shoulder interior/exterior rotation
- Wrist flexion/extension
- Wrist ulnar/radial deviation
- Shoulder horizontal flexion/extension
- Full-range shoulder free motion
- DOF Model:
- Kinematic and Dynamic Ranges of the Human Arm.
- Arm Kinematic Values:
- Act02 = Arm reach to head level
- Act08 = Move object at waist level
- Act10 = Pick up phone on wall/hang up
- Act12 = Eat with Spoon
- Act23 = Eat with Spoon (disabled grasp)
3) From PDF: Kinematic Analysis of Lumbar Spine Undergoing Extension and Dynamic Neural Foramina Cross Section Measurement. By: Yongjie Zhang1, Boyle C. Cheng2, Changho Oh1, Jessica L. Spehar2 James Burgess3 Available on Edge
- Team Looked at the motion of the Lumbar spine of patients that have Lower Back Pain (LBP).
- A Human cadaveric specimen was used to test the flexion extension of the lower back. The team was analyzing an interspinous spacer that was designed to restrict motion primarily in one direction.
4) From PDF: Reduced kinematic model of the human spine. By:Christian Simonidis, Manuel Scharmacher, and Wolfgang Seemann.
- Team looks at a Reduced Kinematic model for the human spine to be used when marker based motion sensors are used to model the motion of the spine. It will describe the motion of the first thoracic vertebra in relation to the pelvis.
- The formulation can offer convenience with sensor placed motion capture, and provide "human-like motion characteristics of torso motion."
- The detailed model of the lumbar spine has any vertebra modeled as a rigid body and the joint axes placed similar to the human spine. The model the team shows has 17 rigid bodies, and 51 degrees of freedom.
- By coupling the between pairs of vertebra the team reduced the model down to three DOF.
- The reduced model is comprised of the following
- R(q1, q2, q3) = rb + AbkAk(q1)[rk + Aker(q2, q3)]
- A(q1, q2, q3) = AbkAk(q1)AkeAe(q1, q2)
- Where R(q1,q2,q3) defines the position vector of the vertebra relative to the pelvis. A(q1,q2,q3) defines the orientation of the vertebra.
- Above q2 and q3 are DOF used for the ellipsoid model, and q1 is axial DOF.
Step 3. Search Internally
Of the prior RIT MSD projects focusing on motion tracking:
- P08006 does not appear to have any information on kinematics or sensor placement in either their technical paper or design review. Their website is incomplete due to it being a spring-fall project, and the spring EDGE page is missing.
- P09027 published information on arm kinematics, including transformation matrices for elbow and wrist coordinates. A screenshot from their final publication is attached below:
- P10007 provides a link to a paper published by Dr. Gombatto titled "Validity and reliability of a system to measure passive tissue characteristics of the lumbar region during trunk lateral bending in people with and people without low back pain." This paper displays much information on the kinematics of a back, for example:
Step 4. Explore Systematically
Specifications of Interest:
- DOF studied for body parts:
- Typical Human BC's/Ranges for body parts: