Son of Toby
Robotic Arm Project
A student research project supported by the Drury University Physics Department , in conjunction with the NASA Reduced Gravity Student Flight Opportunities Program.
The Son of Toby (SOT) project is a continuing physics research project supported by the Drury University Physics Department. Our first robotic arm was flown as an experiment on the NASA weightless wonder aircraft as a part of the NASA Reduced Gravity Student Flight Opportunities Program in the Spring of 2007. The research is conducted by Drury undergraduate students under the guidance of Dr. Greg Ojakangas. SOT is a robotic arm capable of simulating planar human arm motion with six muscles actuating two degrees of freedom. The reason for creating this mechanical arm model is to observe and predict arm motion in the laboratory under normal earth gravity conditions and on NASA's aircraft under near zero gravity conditions, and thus to gain insight regarding how muscle contractions generate arm motions under the vastly different conditions of normal gravity and zero gravity. The arm is designed to approximate some of the dynamical behavior of a true human arm in a microgravity environment. Another purpose of this project is to hopefully assist in understanding how prosthetic arms can be constructed which utilize muscles just as the human body does, in order to help handicapped people.
Our eventual goal is to develop a 4-degree-of-freedom arm as shown here by the matlab simulation. You can find more information about this by visiting Drury University Holodeck.
A Brainstem microcontroller from Acroname Robotics is used to control the servo motions and collect sensor feedback. The major sensors for the original arm were analog potentiometers sampled using the Brainstem 10 bit analog-to-digital (A/D) converters. The arm was also equipped with two accelerometers. These were also sampled using the A/D converters.
Data from our first robotic arm experiment was collected and analyzed. A summary of the results is here. The figure on the right shows one of the trajectories executed by the arm in the airplane. The blue line depicts the circle which was the intended path and the red line depicts the actual path followed by the arm. As you can see, the arm motion was quite crude. There are various reasons including friction and noise in the system behind why the arm could not execute a perfect circle.
The '09-'10 team re-designed the arm model to achieve better results after learning from last year's experience. Emphasis was placed on testing the dynamic response of the system utilizing the new hardware. Initial experiments (shown in image to the right) yielded relatively noise free data with an adequate sample rate (~30 ms). The data shown on the right was used to determine the rotational intertias of the upper arm and the forearm for use in our dynamical equations. A video of a simple free oscillation experiment is here (Note: file is ~2.12MB).
The '09-'10 team designed and constructed a one-degree-of-freedom arm. This simplified laboratory model assisted in characterizing and modeling the dynamic response of the system. As an interesting experimental exercise the 1-degree-of-freedom arm was purposely forced by the two muscles at a frequency different than its natural frequency of oscillation in order to see a chaotic behavior. A description of this behavior is here. The results gained from the simplified one DOF arm were applied to a newly constructed two DOF arm.
Ultimately, we hope the results of our experiments may help better our understanding of human arm motion in the presence of Earth's gravity versus the microgravity enviorment of the "weightless Wonder".