I’m Jim Kerns – the Robotics Lab Instructor. I started here at LTU as an adjunct professor teaching the “Introduction to Mechatronics” class in 2011. Last January, I retired from my former position to take on a full time gig in the robotics lab. Before coming here, I spent more than 36 years at Ford where I mostly worked on powertrain control systems. Looking back, it’s amazing how much the technology and customer expectations have changed since I started working on cars – I can only imagine what kind of technologies that today’s students will be working with 20 or 30 years from now. One piece of advice – never stop learning!
If you have come by the robotics lab (E-6) you will see that it has gone from an empty room to a room full of disassembled robot arms and other equipment. That’s actually a good thing. The robot arms that are currently strewn across the build benches will be up and running in time for experiments in programming and control by this fall. The necessary parts for repair have been identified, and we will be building new driver boxes to allow us to control the arms from either a microcontroller, or the student laptops through the National Instrument Input/Output interface controllers.
Another project was creating some “Segway” like robots using the Lego Mindstorm kits. They can be seen following lines on the central table in the lab or on youtube. The balance control for these is similar to the classic “inverted pendulum” control where you move the base back and forth to keep it under the pendulum center of mass and prevent any rotation (which would be detected by the HiTecnic gyro sensor). But, getting it to move around a track makes the problem a little more interesting; to move one way, you initially have to move the “wrong” way. Let’s consider moving the robot forward. If we were to just drive the wheels forward, the mass of the robot would lag and the center of gravity would end up behind the wheels causing the robot to tend to fall backwards. Instead, we first drive the wheels backwards just a little causing the robot to fall forwards. As it starts to fall, we can then drive the wheels forward to “catch” the fall and start our forward movement. This type of system is commonly referred to as a “non-minimum phase”system.
The robot also uses a light sensor to detect the edge of the white tape on the black table; a simple proportional control is used – if the robot is too far off to the right, the reflected light intensity will be too low and the robot will turn to the left. If it is too far over the line, it turns to the right. A simple, but effective algorithm – up to a point. If it gets past the left edge of the tape, the light intensity drops and it turns left – away from the tape.
Other projects underway include evaluating microcontrollers and devices like this three axis solid state rate gyro. And, we are expecting some samples of “Artificial Muscle” to experiment with.