William T. Tarimo

Quadruped gait learning using cyclic genetic algorithms

Generating walking gaits for legged robots is a challenging task. Gait generation with proper leg coordination involves a series of actions that are continually repeated to create sustained movement. In this paper we present the use of a Cyclic Genetic Algorithm (CGA) to learn gaits for a quadruped servo-robot with three degrees of movement per leg. An actual robot was used to generate a simulation model of the movement and states of the robot. The CGA used the robots unique features and capabilities to develop gaits specific for that particular robot. Tests done in simulation show the success of the CGA in evolving a reasonable control program and preliminary tests on the robot show that the resultant control program produces a suitable gait.

A Flipped Classroom with and Without Computers

Flipping a classroom involves a more interactive class model where students and instructors spend the majority of the class time on various interactive activities in engagement with class materials. Often, this pedagogy style involves taking advantage of the new interactive technologies. In this work, we describe an experiment in an introduction to programming class (CS1) in which we compared the outcomes of offering the same interactive classroom with and without computers. The first approach required students to bring computers to class to engage with the class and materials individually and in groups using the computer-enabled tools. The other approach was to ban computers and require students to interact in person and engage with materials using pen and paper. In both approaches the students’ attempts are shared with the class and discussed, and in general we attempted to maintain the same class models. We found that the use of computers alone had no statistically significant effect on the students’ learning outcomes, enjoyment of the material, self-assessment of their understanding, use of teaching assistant resources, or self-estimate of how many hours they invested outside of the classroom. We did find that a statistically significant number of students preferred in-class engagements and interactions using computers. We also found that the instructor had much more useful and detailed information about individual student’s interaction in class when computers were used. We conclude that, although many instructors are wary of requiring computer use in large classes, there is evidence that students prefer it, it does not negatively affect learning outcomes, and with appropriate tools and pedagogy, it gives the instructor a much deeper and more nuanced view of student performance in the class.

Using Cyclic Genetic Algorithms to learn gaits for an actual quadruped robot

It is a difficult task to generate optimal walking gaits for mobile legged robots. Generating and coordinating an optimal gait involves continually repeating a series of actions in order to create a sustained movement. In this work, we present the use of a Cyclic Genetic Algorithm (CGA) to learn near optimal gaits for an actual quadruped servo-robot with three degrees of movement per leg. This robot was used to create a simulation model of the movement and states of the robot which included the robots unique features and capabilities. The CGA used this model to learn gaits that were optimized for this particular robot. Tests done in simulation show the success of the CGA in evolving gait control programs and tests on robot show that these control programs produce reasonable gaits.

The effects of using a greedy factor in hexapod gait learning

Various selection schemes have been described for use in genetic algorithms. This paper investigates the effects of adding greediness to the standard roulette-wheel selection. The results of this study are tested on a Cyclic Genetic Algorithm (CGA) used for learning gaits for a hexapod servo-robot. The effectiveness of CGA in learning optimal gaits with selection based on roulette-wheel selection with and without greediness is compared. The results were analyzed based on fitness of the individual gaits, convergence time of the evolution process, and the fitness of the entire population evolved. Results demonstrate that selection with too much greediness tends to prematurely converge with a sub-optimal solution, which results in poorer performance compared to the standard roulette-wheel selection. On the other hand, roulette-wheel selection with very low greediness evolves more diverse and fitter populations with individuals that result in the desired optimal gaits.