Gait Generation for Damaged Hexapods using a Genetic Algorithm

Abstract

I would like to thank my parents Sandra Kon and Douglas Kon. For years of support and encouragement of my dreams, as well as your tireless efforts to provide every opportunity imaginable. To my friends. Nicolas Law, Allison Crim, Xavier Tarr, Eli Laramie, Christopher Swider, Derek Bargelt and Nathan Messana. You each have had a tremendous impact on my life and I am so glad to know you all. Thank you for the many years of friendship that have lead me here and I look forward to the many years to come. This would not have been possible with out the endless support, and assistance of numerous wonderful people. Whether you served as a sounding board, a guiding hand or even just lent a supportive ear, know that you helped make this possible. List of Contributions • Design and implementation of simulation model for the hexapod-quadcopter system. • Design and implementation of a Genetic Algorithm for the generation of gaits to compensate for damage in a hexapod system.

Leveraging Evolutionary Algorithms for Feasible Hexapod Locomotion Across Uneven Terrain

“…An interesting problem in locomotion research is creating gaits for "injury" conditions such as complete leg failure (Manglik, Gupta, and Bhanot 2016) or individual joint failure (Kon and Sahin 2020). Similar to our work, Manglik, Gupta, and Bhanot (2016) used combinations of sinusoidal waves to govern all leg servo motion.…”

“…Effective execution of a hexapod gait generally requires a minimum of 18 degrees of freedom. Thus developing an agile and stable gait for traversing randomly uneven terrain has a large search space that serves as a good candidate problem for evolutionary algorithms, and the application of evolutionary algorithms to the field of gait controller optimization has been a subject of ongoing research (Lewis, Fagg, and Solidum 1992;Kon and Sahin 2020).…”