First Two Weeks – Evan Weiler

Wow! How fast two weeks can go by. I feel like I just got off the plane in Minneapolis! Being from a suburb of Los Angeles, the small town of Menomonie has been quite the change of pace from what I am used to, however it has been very welcoming! I’ve settled into the new dorm nicely and have found that the walk to the lab is not as bad as I initially thought it was going to be. There are a few cool places to eat nearby the dorm and the university which is great for when you want to get a quick bite to eat. Overall, I’m liking the area!

Contrary to what I initially expected, we were thrown into the thick of things right away. In our first week, we started by meeting with our team and our faculty advisor, ours being Dr. Cheng Liu. Dr. Liu explained to us his vision for our project this summer, which was to have robotic arm that can be controlled by inputs as seen by a camera. Our goal is to have the camera be able to recognize facial movements and use those movements to control various aspects of the arm. For example, when someone opens their mouth that could be interpreted by the robot and the robotic arm could bring a spoon with food on it to the user.

There are two different ways we can approach this, both involving computer vision. The first is through an FPGA board. This is the method that we have initially begun testing. Currently, we have created a program that can recognize which third of the screen a certain-colored object is and output to one of three bits accordingly. We plan to use this to create a line-following robot, and hope to have this done by the middle to end of this upcoming week. The second method to recognizing facial inputs is through OpenCV which is hosted on Linux through a board called the Beaglebone Black. The advantages of using OpenCV is the fact that there are many open source libraries available. Despite this, working through OpenCV will be slower than working through an FPGA because of the overhead that comes from running Linux. Computations will take longer which is a problem as our project involves real-time inputs. We plan on exploring these advantages and disadvantages in the upcoming weeks.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s