To disseminate advanced knowledge on hot topics in computational intelligence seminars will be organized and distributed through the web free of charge to our members. Past webinars can also be accessed by members free of charge.
Webinar Title: Deep Learning Models for Autonomous Space Guidance
Speaker: Prof. Roberto Furfaro
Webinar Chair: Dr Sansanee Auephanwiriyakul
Date and Time: 2nd March 2020 at 19:00 (GMT – 7)
Registration URL: https://attendee.gotowebinar.com/register/186314849345721357
Autonomous and unconstrained exploration of small and large bodies of the solar system requires the development of a new class of intelligent systems capable of integrating in real‐time stream of sensor data and autonomously take optimal decisions, i.e. decide the best course of action. For example, future missions to asteroids and comets will require that the spacecraft be able to autonomously navigate in uncertain dynamical environments by executing a precise sequence of maneuvers (e.g. hovering, landing, touch‐and‐go) based on processed information collected during the close‐proximity operations phase. Currently, optimal trajectories are determined by solving optimal guidance problems for a variety of scenarios, generally yielding open‐loop trajectories that must be tracked by the guidance system. Although deeply rooted in the powerful tools from optimal control theory, such trajectories are computationally expensive and must be determined off‐line, thus hindering the ability to optimally adapt and respond in real‐time to 1) uncertainties in the unknown dynamical environment; 2) detected hazards; and 3) science value analysis. Over the past few years, enabled by large data availability and advancements in computing hardware (e.g. GPUs), there have been an explosion of intelligent systems based on deep learning that enable adaptive and fast reasoning over data. One can naturally ask the following: how can such techniques help the development of the next generation of robust and adaptive algorithms for space guidance that can learn optimal actions during the course of specified flight missions? In this talk, I will address this problem by presenting a set of deep learning models recently developed by my research team for direct applications to space exploration. The methodologies include the use of Convolution Neural Networks (CNN) Recurrent Neural Networks (RNN) within the framework of deep reinforcement learning and meta‐learning (or “learn‐to‐learn”). The proposed framework enables learning a closed‐loop guidance policy by simulated experience. Such policy (e.g. bank angle, angle of attack as function of the current position and velocity) is parameterized via a deep network and its parameters (weights) learned by experience,
i.e. letting the agent interact with the environment in an attempt to maximize a reward signal. Examples from planetary landing are presented to demonstrate the performance of the proposed approach.
Roberto Furfaro is Full Professor at the Department of Systems and Industrial Engineering, Department of Aerospace and Mechanical Engineering, University of Arizona. He is the Director of the Space Situational Awareness Arizona (SSA‐Arizona) Initiative at the Defense and Security Research Institute (DSRI). He published more than 50 peer‐reviewed journal papers and more than 200 conference papers and abstracts. As PI and Co‐PI, he received more than $30M from NASA and Department of Defense. He was the System Engineering lead for the Science Processing and Operations Center (SPOC) for the NASA OSIRIS REx mission. He is currently the lead for the target follow‐up team of the recently selected NASA NEO Surveyor Mission. In his honor, the asteroid 2003 WX3 was renamed 133474 Roberto Furfaro.
Featured Archived Webinar
Challenging the stigma surrounding the role of women in technology, a journey from combinatorial optimization to IBM
From the IEEE Computational Intelligence Society (CIS), this webinar session presented by Dr Amy Khalfay will cover the role of females within technology and the wider STEM sector.
Many people feel that you must have studied a certain degree, know a programming language, or prefer to work alone to be able to have a career in technology. This is not the case, these careers are open to everyone, from any background. During this session we will be exploring some of the misconceptions about careers within STEM, discovering the many types of roles and doing some myth busting. We will also discuss my personal journey to becoming a graduate technology consultant for IBM, my background of research and my commitment to ensuring more females enter STEM careers. My PhD, titled "Optimization heuristics for solving technician and task scheduling problems", focused on solving NP-hard combinatorial optimization problems that arise in the real world and was sponsored by industry. The project enabled me to enhance my soft skills, write academically, learn to code and develop a deeper understanding of real-world business problems and innovative ways to solve them. Biography: Dr Amy Khalfay is currently a graduate technology consultant for IBM, joining in October 2017. Prior to this Amy completed a BSc in Mathematics (2014) and a PhD in Operational Research (2017). Amy is also a committee member of IEEE Women in Engineering. Amy's research area is combinatorial optimisation solving NP-hard scheduling problems. Areas of skill include Java, Statistical Analysis, Mathematical Modelling and Algorithm Design and Development.
Additional archived Webinars can be accessed here.
When planning to attend a webinar, members should check the location and time zone from where the webinar is broadcast. For all other locations, they should calculate the actual time. Also, members should be aware of the fact that daylight saving time does not begin and end at the same time in different countries.
Members should log in at least 15 minutes earlier to check their connection, hardware, and software and make sure that they are ready to attend the talk when it begins.