Quantum algorithms applied to satellite mission planning for Earth observation

Abstract

Earth imaging satellites are a crucial part of our everyday lives that enable global tracking of industrial activities. Use cases span many applications, from weather forecasting to digital maps, carbon footprint tracking, and vegetation monitoring. However, there are also limitations; satellites are difficult to manufacture, expensive to maintain, and tricky to launch into orbit. Therefore, it is critical that satellites are employed efficiently. This poses a challenge known as the satellite mission planning problem, which could be computationally prohibitive to solve on large scales. However, close-to-optimal algorithms can often provide satisfactory resolutions, such as greedy reinforcement learning, and optimization algorithms. This paper introduces a set of quantum algorithms to solve the mission planning problem and demonstrate an advantage over the classical algorithms implemented thus far. The problem is formulated as maximizing the number of high-priority tasks completed on real datasets containing thousands of tasks and multiple satellites. This work demonstrates that through solution-chaining and clustering, optimization and machine learning algorithms offer the greatest potential for optimal solutions. Most notably, this paper illustrates that a hybridized quantum-enhanced reinforcement learning agent can achieve a completion percentage of 98.5% over high-priority tasks, which is a significant improvement over the baseline greedy methods with a completion rate of 63.6%. The results presented in this work pave the way to quantumenabled solutions in the space industry and, more generally, future mission planning problems across industries.

Hybrid Quantum Neural Network for Drug Response Prediction

“…Since the existing classical models require significant computational resources, this limits their performance. Quantum and quantum-inspired computing models can potentially improve the training process of existing classical models [12][13][14][15][16][17][18], allowing for better target function prediction accuracy with fewer iterations. Some of these methods provide a polynomial speedup, critical for large and complex problems where small improvements can make a big difference.…”

Benchmarking Simulated and Physical Quantum Processing Units Using Quantum and Hybrid Algorithms

“…Quantum neural networks (QNNs) are quantum machine learning (QML) algorithms [1][2][3][4] that leverage powerful techniques developed for classical neural networks, to optimize this parameterized structure, and have already been applied to solve a number of industrial problems. [5][6][7][8][9][10][11][12] The complexity and performance of classical neural networks employed to solve data-intensive problems has grown dramatically in the last decade. Although algorithmic efficiency has played a partial role in improving performance, hardware development (including parallelism and increased scale and spending) is the primary driver behind the progress of artificial intelligence.…”

Parallel Hybrid Networks: an interplay between quantum and classical neural networks

“…Therefore, contemporary practical use of quantum technologies in machine learning should come from complementary quantum-classical architectures, called hybrid quantum neural networks (HQNN), that employ relatively small, realisable quantum circuits and classical multi-layered perceptrons (MLP) where the two work in tandem. In [7][8][9][10], we explored the applicability and performance of sequential HQNNs, where MLPs and QNNs are connected in series, passing the information from one network to another. The sequential HQNNs could introduce information bottlenecks in the representational power of the model, which could limit the expressivity of the network.…”