Force Measurement Using Optical Tracking
Internship with Smith & Nephew
Summer 2020
Final 3D Printed Test Device
Initial Sketch
CAD Model
For the summer of 2020, I designed and completed a study to determine if accurate force measurement using camera tracking is feasible. I took this project from initial ideation to an engineering report as a way to publish the results.
​
The picture to the left shows the final test device that I designed. The geometric shapes with the blue markers are trackers that a camera can segment and identify. Since the goal was to use cameras to measure force, I wanted to use two trackers to measure the change in displacement.
​
The design revolves around using springs with known spring constants. In the sketch below, you can see the placement of two springs: one linear spring, and one torsional spring. By using Hooke's law, I can measure the displacement and then multiply by the spring constant to find a measured force.
​
Due to COVID-19 restrictions, this internship was remote, lending to multiple challenges. Primarily, I had no access to the regular prototyping tools that I usually have during a design process. This design was going to be 3D printed by someone else, so before finalizing the design, I had to ensure the tolerances were near perfect, so that multiple prints wouldn't be necessary. Through this project I learned a lot about how to properly tolerance a CAD model. Furthermore, since I was unable to go to the office, I wrote a testing protocol, but was unable to complete the testing. Someone else followed my protocol, and sent me the data to analyze.
​
By using a force gauge to measure the actual force during testing, after analyzing the camera data, I was able to compare the ground truth force with the measured force. More on this below.
On the left, is a snapshot of the unintelligible raw camera data. Using Octave, I wrote a script to convert the data into easily readable graphs, so that I can easily see tracker displacement and the corresponding force. On the bottom graph in the right side picture, the point marker shows the read out of the force gauge during testing (the ground truth force) while the blue line shows the measured force based on tracker displacement. The only point of interest on the blue "Force Measured" line is at the peak, since force is applied slowly until a desired force is reached, at which point during testing, the force gauge measurement was recorded. As it can be seen, this test device yielded accurate readings. The results ultimately showed that using camera to measure force was feasible, and worthy of more in depth research. Furthermore, while I designed this test device to measure both force and torque, due to time constraints, only compression was tested.