Team AERO strives to foster a dedicated and collaborative team, composed of graduate and undergraduate students, to develop and implement innovative open-source control, navigation, perception, and manipulation algorithms enabling robots and humans to reach further into space. We aim to demonstrate successful autonomous collection of geologic samples of interest utilizing AERO with the intent of gaining experience in the holistic system design process, improving our robotics engineering skills, and building a kick a** robot.
The insatiable human curiosity and desire to explore proved Earth is neither the sole planet in the solar system nor the center of the universe. Technologies developed over the last 40 years have significantly enhanced humanity’s ability to explore extraterrestrial bodies with teleoperated robots returning stunning images and scientific data of distant planets. This research aims to develop AERO, the Autonomous Exploration Rover, to advance the next evolution in space exploration to go where teleoperated robots cannot bring us. AERO is comprised of a differential-drive four-wheeled mobility platform and 6-DOF manipulator designed to participate in the NASA Sample Return Robot Centennial Challenge. The task is to navigate a large outdoor area, find and locate various samples, and return them to the starting platform. Fusing a combination of data from a fixed, forward-facing stereo vision system, LIDAR, and IMU, AERO implements a simultaneous localization and mapping (SLAM) algorithm to mark what areas are searched and return to the starting platform at the end of the competition. A second panning stereo vision system on a mast is used to locate and identify samples using object classifier and texture-based algorithms. Because the time-delays for teleoperated robots beyond the moon are unacceptably long, the algorithms developed to identify samples defined by high-level descriptions will allow humans to efficiently explore distant celestial bodies. Finally, these algorithms with slight modifications can solve problems on Earth including agricultural robots identifying crop diseases, security robots identifying security breaches, and home-care robots identifying household risks to seniors.
More info about the competition can be found here: NASA Sample Return Robot Challenge