Research/ PLANETARY ROBOTICS
Robust Machine Learning Systems For Dependable Space Applications
Nickolaos Panagiotopoulos, Toni Lubiniecki, Alen Turnwald, Niklas Baldauf
Learning-based MethodsOpenAlex topic: Reinforcement Learning in Robotics · Perception & State EstimationOpenAlex topic: Robotics and Sensor-Based Localization · Motion Planning & NavigationOpenAlex keyword: rover navigation
Artificial Intelligence and Machine Learning have attracted lots of attention to the space engineering community the last few years. In this paper, an attempt to utilize the concept of Machine Learning is presented for a robotic mission to other planets (in this paper is focused on Mars) for planetary exploration purposes. Although, building a machine that is able to learn is not achieved (and it is not yet feasible to be built in general), the borrowed ideas, concepts, and products from the research domain on Machine Learning, provided the necessary engineering tools for improving the navigation, kinematics, and control of the rover.