Invited Speakers
Exobiology Extant Life Surveyor (EELS) concept. Credit NASA/JPL-Caltech
Friday, July 7, 2023
9:00 - 10:00
Autonomous Task and Motion Planning for Ice Worlds Exploration
Tiago S. Vaquero, Guglielmo Daddi, Rohan Thakker, Mike Paton, Ashkan Jasour, Mitch Ingham, Hiro Ono
Jet Propulsion Laboratory, California Institute of Technology
Abstract:
Icy moons of the gas giant planets, such as Europa, Titan, and Enceladus, are at the forefront of astrobiological interest due to the evidence of subsurface oceans. Enceladus in particular is unique among the icy moons because there are known vent systems that are likely connected to a subsurface ocean, through which the ocean water is ejected to space. An existing study has shown that sending small robots into the vents and directly sampling the ocean water is likely possible. To enable such a mission, JPL is developing a novel, snake-like robot called Exobiology Extant Life Surveyor (EELS) that can navigate Enceladus' extreme surface and descend an erupting vent to capture unaltered liquid samples and potentially reach the ocean. However, navigating to and through Enceladus' environment is challenging: due to the limitations of existing orbital reconnaissance, there is substantial uncertainty with respect to its geometry and physical properties of the surface/vents; communication is limited which requires highly autonomous robots to execute the mission with limited human supervision. In this talk we provide an overview of the EELS project and its development effort to create a risk-aware autonomous robot to navigate these extreme ice terrains/environments. We describe the robot’s architecture and the technical challenges to navigate and sense the icy environment safely and effectively. We focus on the challenges related to surface mobility, task and motion planning under uncertainty, and risk quantification. We provide initial results on mobility and risk-aware task and motion planning from field tests and simulated scenarios.
Speaker Bios
Tiago Vaquero is a Technical Group Lead in the Artificial Intelligence, Integrated Planning and Execution Group at the Jet Propulsion Laboratory, California Institute of Technology. He holds a B.Sc., M.Sc., and Ph.D. in Mechatronics Engineering from the University of Sao Paulo, Brazil. Tiago previously held a MIT research scientist position and a joint Caltech/MIT research position where he worked on Resilient Spacecraft Systems and Risk Sensitive Planning/Scheduling algorithms. At MIT, Tiago also worked on Risk-aware Planners and Executives for Autonomous Underwater Vehicles and Autonomous Cars. Tiago also previously held a research position at the University of Toronto where he worked on Multi-Robot Planning and Coordination. He current leads the development of uplink tools in the context of Operations for Autonomy and leads the mission planning team in the JNEXT EELS project. His research interest includes knowledge engineering for autonomous vehicles, and automated planning and scheduling for single and multi-robot missions.
Guglielmo Daddi is an aerospace Ph.D. student at Politecnico di Torino. He focuses his research on system-level autonomy and works with the autonomy team of JPL's EELS project. At JPL, he also worked on modeling and prototyping a plasma-discharge-enhanced melt probe for planetary subglacial access. At Politecnico, he worked on simulation environment development and payload systems engineering for ESA-SROC Phase B and other small sat concept assessments.
Rohan Thakker is a Robotics Research Technologist at NASA’s Jet Propulsion Laboratory. He is the Autonomy Lead of the Exobiology Extant Life Surveyor (EELS) project. Previously, he led JPL’s Guidance, Navigation, and Controls team, COSTAR, for DARPA Subterranean Challenge at the Tunnel Circuit and Short-range Planning and Learning team for DARPA RACER. His research focuses on developing resilient full-stack autonomy capabilities by tightly coupling planning, control, and estimation algorithms. He has deployed these algorithms on drones, legged robots, tracked/wheeled rovers, off-road cars, and snake robots in real-world environments. Rohan has been working at JPL since 2017 after finishing graduate studies in robotics at Carnegie Mellon University.
TARANTULA NEBULA This image of the Tarantula Nebula captured by JWST and released by NASA on Sept. 6, 2022 spans 340 light-years across. The observatory's infrared detectors revealed a cluster of never-before-seen young stars at the center of the image that were previously shrouded by dust.Image: NASA, ESA, CSA, STScI, Webb ERO Production Team
Saturday, July 9, 2023
13:30 - 14:30
Planning and Scheduling the James Webb Space Telescope: First-Year Experiences and Challenges
Mark Giuliano, Reiko Rager and Nazma Ferdous
Space Telescope Science Institute (STScI)
Abstract
The James Webb Space Telescope was launched on December 25th, 2021, and was inserted into orbit around the second Sun-Earth Lagrange point on January 21st, 2022. After 6 months of commissioning the James Webb Space Telescope (JWST) began science operations on July 11th, 2022. The SPIKE planning and scheduling software system as initially developed for the Hubble Space Telescope (HST) is used for JWST long range planning. We review the observation planning and scheduling process flow for JWST showing how the approach extends the process used for HST. We highlight the challenges faced by the development team over the first 9 months of operations including
Post launch it became apparent that mission defined parameters would need to be updated frequently and that an architecture which required recalculating constraints, even when window constraint parameters did not change, for each update was not sufficient.
Early in the mission the JWST mirror was hit by micro-meteoroids with a frequency significantly higher than expected. We describe how SPIKE’s ability to flexibly add constraints allowed operations to segue from engineering studies, to a prototype, to an operational system which minimizes the chance of future meteoroid strikes.
A high percentage of JWST observations are linked to other observations via temporal constraints (e.g., Observation 2 after Observation 1 by 10-20 days). When an observation is executed on board the exact execution time is fed back into the long-range planning system to adjust the constraints of linked observations. Unlike HST, many JWST observations are either executed outside of calculated windows or are skipped entirely. We describe how features of the SPIKE temporal constraint propagator that have been in place for over 30 years of HST operations were extended to handle such anomalies.
A grating wheel for the JWST MIRI instrument was found to have excess friction when switching directions and scheduling of these observations was became highly constrained. The team is currently developing software to mitigate the problems with the wheel by having the planning system define campaigns which control the frequency of direction changes and ensure that the wheel operates within safe limits.
In order to ensure quality across the entire mission, the software development process for JWST is more formal than the iterative prototype driven approach we have used for HST. We describe the development process and its pros and cons.
Speaker Bios
Mark Giuliano is a Consulting Software Engineer and manager of the Astronomical Planning and Support Branch at the Space Telescope Science Institute (STScI). Dr Giuliano received his Ph.D. in Computer Science from the University of Maryland. Mark has over 32 years of experience at STScI and is responsible for the design, development, and maintenance of the SPIKE planning and scheduling software used in operations on the Hubble, James Webb, Chandra, and the upcoming Roman Space Telescope. His expertise includes system architectures to support the preparation, planning, and scheduling of space-based observatories and the design of planning systems.
Reiko Rager is a Principal Software Engineer in the Astronomical Planning and Support Branch at the Space Telescope Science Institute (STScI). Dr Rager received her Ph.D. in Computer Science from the University of Maryland. Reiko has over 24 years of experience at STScI and has worked on the design, development and maintenance of planning and scheduling subsystems for the Hubble, James Webb and the upcoming Roman Space Telescope. Dr Rager’s expertise includes mathematical constraint modeling, distributed computing, and project management. She leads the development of the software system that enables parallel science for the Hubble Space Telescope and the software systems that check scheduling feasibility for the James Webb and Roman telescope observation preparation systems.
Nazma Ferdous is a Principal Software Engineer in the Astronomical Planning and Support Branch at the Space Telescope Science Institute (STScI). Dr Ferdous received her Ph.D. in Computer Science from City University of New York, Graduate Center. Nazma has over 22 years of experience at STScI and has worked on the design, development and maintenance of planning and scheduling subsystems for the Hubble, James Webb, Chandra, and the upcoming Roman Space Telescope. Her expertise includes software design and knowledge engineering. She is responsible for creating and maintaining the heuristic systems that address the inherent computational complexity of observation planning for astronomical observatories.