Mathematical Analysis-Thursday March 30th, 2017

This is the equation for Garg's Method! But don't worry its less intimidating than it looks. The first section of the equation requires the summation of the energy of position multiplied by time. Position is basically standing or sitting. Since it takes certain energy to stand or sit this must be factored into the overall energy for the job. For both squats, the position is standing. The second part of the equation requires the summation of the change in energy of all tasks. For the lifts in my research this can be broken down to two motions: the lift and the arm raise. Since I lifted both boxes to the same height, the energy used in the arm raise will be the same for both the stoop and squat lift. The lift motion accounts for the energy until the load reaches 0.81 meters or hip height. Once the load reaches that height the arms begin lifting the load to its final height(in this case 1.42 meters). Calculating the energy for the squat and stoop lift requires two other equations that are impacted by initial height, final height, body weight, load weight, and sex(male or female). The equation for arm raise requires a very similar equation, but it isn't necessary here because it is the same for both lifts.

In my case, initial height was zero because I lifted the load from the ground. For these two equations, final height was 0.81 because I lifted the load above my hips. The variable S represents sex. Insert a 1 for a male subject and 0 for a female subject. The box that I lifted weighed 20 kg. I found time of lift by looking at how many frames of data were captured by the Vicon system. The stoop lift took longer at 353 frames or 3.53 seconds. The squat lift ended in 3.19 seconds. After calculating this all out, I found that the squat lift requires 18.629 kilocalories per minute while the stoop lift only requires 13.139 kcal/min. It is interesting to note that the stoop lift requires less energy despite being considered the inferior lift, but more on this next week!

Garg's Method Introduction-Monday March 20th, 2016

First off, sorry for the late post. I wasn't able to travel to Prescott last week because of a spring snowstorm. I was supposed to start Garg's method last week. Garg's method is a way to "estimate metabolic energy expenditure rate for a job." This is will be the majority of my physiological analysis. This measures the rate at which energy is used for a task. Garg's method will normally be used in ergonomic research where a worker has to perform a task repeatedly throughout the day. This rate of energy production is measured in kilocalories per minute. For reference, an acceptable level of exertion for a worker over the course of a day is 5.2 kcal/min. For the next couple days, I will continue researching Garg's method and calculating the energy expenditure for my data. I will post my findings once I've had the opportunity for my advisor to review my calculations.

I also need to correct something from my last post. The program that I was using to graph and run profiles on my data wasn't actually using all the data. This means the graphs I posted in my last entry are wrong. Here are the corrected versions!

Squat:

Stoop:

Static Equilibrium-Friday March 10th, 2017

This week I transitioned from working with the Vicon system, to researching what is actually happening during the lift. This involved looking at two things to start: physics, which I am very familiar with, and excel, which I am not familiar with. Using excel, I had to convert the units for my data, and organize all my data. There are more than 3,600 data points for just one of my lifts. After organizing the data, I transferred it into an analysis program. The analysis program showed us a bunch of information and graphed the analysis for me. We got some puzzling results, so we're still trying to make sure this is accurate. The first graph is the squat lift, the second is the stoop. As you can see, both are very similar and flat. What is puzzling is how flat both graphs are.

After this, I began analyzing my data for just a single frame of the data. This involves assuming a static equilibrium in order to solve for the force and torque at the elbow, shoulder and hip. By assuming a static equilibrium, all the forces and torques must cancel out to equal zero. This is something that I am familiar with doing. One new step that had to be taken was using a proportion to find the approximate weight of arm and wrist, as well as the length. Instead of trying to measure these things by hand, they can be approximated after only knowing the subject's height and weight. For example, take the height of the subject in centimeters and multiple it by (0.146 + 0.108) to find the length from elbow to hand. Because I'm 6'6"(about 198 centimeters) this value was 36.828 centimeters in my calculations. The last thing that must be found is the angle, relative to the horizontal, at the hip, shoulder and elbow. This can be found using the inverse of tangent and the x,y coordinates of the wrist, elbow, and shoulder. By taking the inverse tangent of the difference of the y coordinates, divided by the difference of the x coordinates, the angle can be found. This angle needs to be found in order to solve for torque. Next week, is spring break so I will not be posting, but the week after that, I'll start looking at what this information means, and what to do with it.

33 Point Model! Thursday March 3rd, 2017

This week I had a chance to work with the 33 point model which was a huge step up from the 6 point model. The detail achieved with this model was incredible, but it came with its own slew of problems. Starting from the beginning, we first struggled with identifying where all the markers would go. We ended up with seven markers on each arm, six on each leg, four on the chest, and three on the head. The frustrating part was the double sided tape wasn't sticking very well to the suit, so after we decided where to put all the markers they kept falling off. Sort of like trying to keep two dogs in a bath tub, except there were 33 of them. Next, identifying markers in the system became a lot more difficult, like an intense game of connect the dots. Unfortunately, because of the cameras were positioned, it was hard to have all the markers show up during the capture. If one of the markers was missing, it wouldn't become apparent until the capture was completed. This meant it took us at least four or five attempts at subject prep to identify all the markers. Once this was completed, we were able to perform a capture. The majority of our time was then spent making sure all the markers were properly labelled and there were no gaps in the data. This happens with the six point model, so I was more familiar with this. We only did one lift, the squat lift, with this model because of the number of other problems that we encountered.