Software & Computing Systems
Intelligent and Autonomous Systems
Intelligent and autonomous software technologies are arising that allow spacecraft to use knowledge and reasoning procedures to determine, for themselves, actions that need to be taken to meet their mission goals. New generations of satellites, rovers, and spacecrafts are able to monitor their own health and update actions and procedures.
As humankind extends its robotic presence throughout the solar system and even beyond, a new breed of autonomous spacecraft is required. The communication delay inherent in deep-space travel makes direct human control less and less practical as distances increase and missions grow more ambitious. Robotic explorers need to be able to respond intelligently to unexpected events, recognize and avoid hazards, seek out science opportunities and recover robustly from faults. Greater spacecraft autonomy and ground-system automation reduces mission operations costs.
JPL is conducting research in several areas of autonomy technology, including integrated mission planning and execution, and fault protection, and onboard science.
Selected Research & Ongoing Projects
High-Reliability Onboard Machine Learning Methods
This task developed software to detect and correct radiation-induced errors in computation performed onboard spacecraft. If the software can detect and respond to single-event upsets (SEUs), reducing the need to harden the hardware against radiation, the range of potential components that can be used on spacecraft widens dramatically. Results showed that fault-tolerant Support Vector Machine classification or regression could be reliably used onboard a spacecraft, albeit with a high false-alarm rate. The task also developed “BITFLIPS,” a software simulator for testing fault-tolerant machine-learning methods in a variety of radiation environments.
CLARAty
Coupled-Layer Architecture for Robotic Autonomy (CLARAty) is a software framework that promotes reusable robotic software. It was designed to support heterogeneous robotic platforms and integrate advanced robotic capabilities from multiple institutions, so its design is portable, modular, flexible and extendable. CLARAty supports the development and integration of advanced robotic technologies under the Mars Technology Program and other NASA programs. It is a collaborative effort among JPL, NASA Ames Research Center, Carnegie Mellon, and the University of Minnesota.
|
|
Sparky is one of the rover test beds that is used to develop, integrate, and test rover technology algorithms and related CLARAty-compliant software.
|
The Machine Learning and Instrument Autonomy (MLIA)
MLIA is composed of a group of researchers that create software solutions to problems requiring data mining, knowledge discovery, pattern recognition, and automated classification and clustering. The underlying emphasis is on building systems based on learning algorithms. Their focus is on the automated analysis of scientific data generated by NASA and JPL instruments, on the development of technologies for adaptive systems, and on enabling technologies for autonomous spacecraft. Examples of MLIA projects include the Multi-angle Imaging SpectroRadiometer (MISR), bioinformatics tools for human chromosome HSA 19, and the Onboard Autonomous Science Investigation System (OASIS).
MISR captures images of Earth at moderate resolution (275 m or 1.1 km) from nine different angles. By comparing images of the same area from different angles, scientists are able to identify thin clouds and determine approximate cloud heights with unprecedented accuracy, leading to greater understanding of the planets global distribution of clouds and how that affects the global climate. Automating the process of detecting clouds and distinguishing between different types of clouds and aerosols remains a challenge. The MLIA Group is applying machine learning technology to this problem to complement the physics-based algorithms currently in use.
The MLIA Group is developing a variety of bioinformatics tools to support the efforts of biologists at Caltech and Lawrence Livermore National Laboratory to produce a complete functional annotation of genes and regulatory sequences in the unusually gene-dense human chromosome HSA19. The tools include automated image-analysis software that enables high-throughput interpretation of tissue arrays, and systems for integrated analysis of diverse sequence annotation and transcript-expression data.
OASIS is an onboard technology which evaluates the geologic data gathered by rovers such as those employed to explore Mars. This analysis is used to prioritize the data for transmission to Earth and to identify science opportunities, which the rover can exploit thanks to the system’s planning-and-scheduling component.
Artificial Intelligence Group (AIG)
|
|
AIG performs basic research in the areas of artificial-intelligence planning, scheduling, and execution with applications to science analysis, spacecraft commanding, deep-space-network operations, and space transportation systems. Examples of AIG’s projects include the Autonomous Spacecraft Experiment (ASE), the Earth Observing Sensorweb, and rover autonomy such as OASIS.
ASE has been operating onboard the Earth Observing One mission since 2003. The ASE software uses onboard continuous planning, robust task and goal-based execution, and onboard machine learning and pattern recognition to radically increase science return by enabling intelligent downlink selection and autonomous retargeting. This software demonstrates the potential for space missions to use onboard decision-making to detect, analyze, and respond to science events, and to downlink only the highest-value science data. See the featured publication, “Using Autonomy Flight Software to Improve Science Return on Earth Observing One” on the AIG publications page.
The Earth Observing Sensorweb uses a network of sensors linked by software and the internet to an autonomous satellite observation response capability. This system of systems is designed with a flexible, modular, architecture to facilitate expansion in sensors, customization of trigger conditions, and customization of responses.
|
The Autonomous Sciencecraft Experiment (ASE) uses onboard science analysis and replanning to radically increase science return via intelligent downlink selection and autonomous retargeting.
|
Strategic Assessment of Risk and Technology (START)
START teams conducted two rover-autonomy studies to help determine the relative benefits of investing in various software technologies that could enable Mars Science Laboratory (MSL) to do science, avoid most failures, and diagnose and correct their own problems with less need to phone home for help.
|
|
Increased autonomy could enable Mars Science Laboratory (MSL) to increase both efficiency and productivity and could enable future clusters of rovers to work cooperatively.
|
|
Contacts
Larry Bergman - Management Contact
E-Mail: Larry.A.Bergman@jpl.nasa.gov
Phone: 818.393.5314
Ben Smith - Technical Contact
E-Mail: Benjamin.D.Smith@jpl.nasa.gov
Phone: 818.393.5371
Links
