Post #17

May 27, 2025

#17a: Stretch Movement (Jonathan)

Adjusting where the stretch goes to during a therapy session and how it responds to user movement is the most important facet of our project. This far into our implementation, we have created an ArUco marker to be detected by the Stretch camera as well as programmed some of the Stretch's movements around this marker. As this video shows, our stretch robot localizes itself in regard to the chair with the marker placed on it. The robot then cycles through a routine of poses, determined by moving the gripper position to various angles. We have chosen four position directions (the arm at 0 degrees, 90 degrees, 180 degrees, and 270 degrees) that will be the most important measurement points during the session. Future work includes increasing the range of rotation (we kept it low for testing purposes), adding positioning for leg movement, and exploring autonomous detection of chair and ArUco marker.

#17b: User Interface (Luna Frontend, Deven SQL)

Our user interface also made significant progress for displaying results and supporting a variety of actions. We have created our website using node and have multiple pages for logging in/registering. By extension, we also have created a backend relational database using mySQL to record users and potentially their progress over time (still discussing this). We have also made improvements to the session page so that the user can manually operate the robot in the case that they want to conduct more specialized training. We still need to add the buttons specifying arm or leg weakness as well as begin collecting force data from the sensor to be displayed on the graph. We plan on creating this force gauging graph using a React chart library.

(This UI text needs to be updated!!)

#17c: Scale Integration (Vlad)

Measuring force from our external sensor is the final main component of our project. We have researched various types of scales and have settled on a standard kitchen scale supporting up to 30 lbs (which should be well above the range of force supplied by patients). We have utilized a c-based library for communicating sensor readouts and have tested in on the Stretch robot. We have successfully received the sensor output after ssh'ing into the robot and must now remotely communicate it to our backend to be displayed on the live graph.

Page updated

Google Sites

Report abuse