Complementing humans in space.
About autonomy and robotics
[image-62][image-67]Future space exploration missions require complex vehicles, habitats, robotic assistants and self-sufficient spacecraft systems which adapt to complex, rapidly changing environments.
Additional exploration technologies for ground and flight operations must include automated planning and scheduling to increase the safety of these missions and reduce their cost.
Similarly, automated planning is crucial in order to maximize science return from deep space probes and even terrestrial observing systems, and to complement and enhance the capabilities of humans doing mission operators.
Since 1990, the NASA Ames Intelligent Robotics Group (IRG) has led the agency in developing robotics technology to reinvent planetary exploration. Ames develops and integrates new technologies into autonomous systems for flight missions and terrestrial demonstrations.
Areas of research and development include:
- Adaptive control technologies
- Control agent architectures
- Embedded decision systems
- Evolvable systems
- Intelligent robotics
- Adjustable autonomy
- Distributed and multi-agent systems
- Goal-level commanding
Planning and scheduling
Featured example: Intelligent robotics
How is the NASA Ames Intelligent Robotics Group reinventing planetary exploration?
[image-83]Since 1990, the NASA Ames Intelligent Robotics Group (IRG) has been developing robotics technology to reinvent planetary exploration. A key part of our work is to field test advanced mobile robots for exploration. These robots perform work that cannot, or should not, be carried out by humans. Some of this work will be performed in parallel with humans, but many tasks can be done before and after humans are present.
To support mission planning and remote science, IRG is developing automated planetary mapping systems. These systems are needed to process the enormous amount of satellite data that NASA collects with spacecraft such as the Lunar Reconnaissance Orbiter, as well as the vast stores of planetary data from previous planetary missions. In our work, we make extensive use of computer vision and cloud/supercomputing.
IRG is actively involving the public in space exploration with neo-geography tools and on-line collaboration such as Mars in Google Earth and Moon in Google Earth. We also developed the GigaPan robotic camera We also work closely with Carnegie Mellon University to develop the GigaPan robotic camera, which we use for citizen science.
Featured example: Planning and scheduling
How does Ames use its long history of R&D in cutting edge automated planning to address NASA’s technical challenges, in addition to infusing this technology into a wide range of NASA missions?
[image-99]Future exploration missions to the moon and Mars will involve complex vehicles, habitats, and robotic systems. Automated planning and scheduling will increase the safety of these missions and reduce their cost. Similarly, automated planning is crucial in order to maximize science return from deep space probes and even terrestrial observing systems. Finally, automated planning complements and enhances the capabilities of human operators.
In collaboration with other groups at the Center Ames as well as other parts of NASA, the Planning and Scheduling Group has developed reusable automated planning and plan execution technologies for use in NASA missions. These technologies include EUROPA, a toolbox for building automated planning systems, and PLEXIL, an automated plan execution technology. In addition to these technologies, the Planning and Scheduling Group has developed advanced prototypes of automated planning systems for many other applications across NASA’s mission directorates.
Featured example: SPIFe
How does an Ames-developed computer program allow scientists on missions like MSL Curiosity deal with planning and scheduling constraints and conflicts?
[image-115]SPIFe has played a key role in a wide variety of NASA missions and terrestrial analogs and simulations, ranging from daily science planning for the Mars Science Laboratory mission to crew and ground scheduling for the International Space Station.
The SPIFe toolkit was designed from the ground up with the needs of the mission operations user in mind, and it presents unique solutions to a number of problems common in other commercial and homegrown systems. Many planning tools are developed first as user-facing interfaces to automated planning systems, and do not allow users enough flexibility to explore plans in a number of different ways, quickly understand complex sets of constraints and their implications, or experiment with different solutions without fear of losing work. SPIFe was built to address those issues in a package that is usable directly by mission scientists to understand critical resource tradeoffs to maximize return from a given ops situation. SPIFe is an integrated planning and scheduling toolkit based on hundreds of hours of expert observation, use, and refinement of state-of-the-art planning and scheduling technology.
The SPIFe user interface is designed to be a highly adaptable and user-customizable framework for viewing and manipulating plan and schedule data. In order to achieve this, SPIFe employs a composable, plug-in architecture based on the the open source Java Eclipse Rich Client Platform (RCP). Eclipse provides a robust plug-in framework, and the RCP provides many fundamental user interface components, such a tabbed ”workbench” that allows users to manipulate views and editors to display the information most relevant to the task at hand.