Machine-Insect Interface: Spatial Navigation of a Mobile Robot by a Drosophila

• Main Author: Shivakumar, Suddarsun
• Format: Others
• Published: OpenSIUC 2018
• Online Access: https://opensiuc.lib.siu.edu/theses/2324; https://opensiuc.lib.siu.edu/cgi/viewcontent.cgi?article=3338&context=theses

Machine-insect interfaces have been studied in detail in the past few decades. Animal-machine interfaces have been developed in various ways. In our study, we develop a machine-insect interface wherein an untethered fruit fly (Drosophila melanogaster) is tracked to remotely control a mobile robot. We develop the Active Omni-directional Treadmill (AOT) model, and integrate into the mobile robot to create the interface between the robot and the fruit fly. In this system, a fruit fly is allowed to walk on top of a transparent ball. As the fly tries to walk on the ball, we track the position of the fly using the dark field imaging technique. The displacement of the fly will be balanced out by a counter-displacement of the transparent ball, which is actuated by the omni-directional wheels, to keep the fly at the same position on the ball. Then the mobile robot spatially navigates based on the fly movements. The Robotic Operating System (ROS) is used to interface between the ball tracker and the mobile robot wirelessly. This study will help in investigating the fly’s behavior under different situations such as its response to a physical or virtual stimulus. The future scope of this project will include imaging the brain activity on the Drosophila as it spatially navigates towards a stimulus.