NASA Embarks on the Quantum Computing Path

Overview

NASA has recently embarked on a journey to explore the potential for quantum computers to tackle problems that are difficult or impossible for traditional supercomputers to handle. The core of many real-world problems consists of extremely difficult combinatorial optimization challenges that cannot be resolved accurately by classical supercomputers in a realistic timeframe. Today’s supercomputers are not designed to effectively solve problems in areas such as image classification, machine autonomy, verification and validation, and planning and scheduling. In contrast, quantum computers use a physical evolution process that harnesses quantum mechanical effects such as entanglement, tunneling, and wave-particle duality. NASA researchers will investigate areas where quantum computing and quantum algorithms have the potential to dramatically improve the agency’s ability to solve difficult optimization challenges in aeronautics, Earth and space sciences, and space exploration.

Project Details

The NASA Advanced Supercomputing (NAS) Division at Ames Research Center hosts the Quantum Artificial Intelligence Laboratory (QuAIL), a five-year collaborative effort among NASA, Google, and Universities Space Research Association (USRA). The NAS facility houses a 512-qubit D-Wave Two quantum computer, currently the largest such system in the world. The facility has been extensively retrofitted to provide isolation from vibration and Earth's magnetic flux, as well as the infrastructure required to cool the system to its near-absolute-zero operating temperature.

For certain types of optimization problems, the quantum properties of a qubit will allow QuAIL researchers to effectively try all possible solutions at once, conceivably arriving at the best answer much more quickly than is currently possible using existing supercomputers. Applications relevant to NASA include machine learning, pattern recognition, mission planning and scheduling, distributed navigation and coordination, and system diagnostics and anomaly detection.

Results and Impact

The D-Wave Two system has been operational since September 2013, with the first results expected in early 2014. So far, QuAIL researchers have developed benchmark sets of hard optimization problems in the context of NASA mission planning and scheduling. These sets are parameterized to generate problems in a range of sizes that allow us to compare the effectiveness of different quantum annealing approaches and various state-of-the-art classical planners. We have developed four different QUBO (quadratic unconstrained binary optimization) representations of these planning problems—the form of input the D-Wave system understands—to enable us understand the best way to leverage quantum annealing. Additionally, we have formulated a mapping to the QUBO form for diagnosing multiple faults, such as those found in the electrical power systems of aircraft and spacecraft. Benchmark examples of these fault management problems are being prepared to run on the quantum hardware. This work will be documented in a series of technical reports and publications in late 2013 and early 2014.

Role of High-End Computing Resources

Quantum computing is the next frontier of high-performance computing. In fact, the first generation of quantum computers will operate as special-purpose processors attached to traditional supercomputing systems. This is one of the primary reasons that the D-Wave Two system is located within the NAS facility, which houses Pleiades, NASA's primary supercomputer.