Beyond the Algorithm Challenge

We propose a novel solution titled Quantum Hyperdimensional Computing for Surface Water Signatures, which integrates quantum-inspired encoding with hardware-friendly symbolic AI to address challenges in flood mapping. Our approach combines hyperdimensional computing and quantum processing to create robust, noise- and haze-resilient vector representations of satellite data, including SAR and optical modalities. These encodings support single-pass, backpropagation-free learning using binary operations, enabling fast and energy-efficient model deployment on edge hardware. By fusing multi-channel and multimodal inputs into high-dimensional vectors, our method ensures accurate and scalable classification of surface water signatures even under extreme environmental conditions. We demonstrate superior performance across standard datasets using custom quantum-enhanced pipelines and ASIC simulations. The proposed architecture targets both scientific insight and deployability, bridging the gap between unconventional computing and real-world flood resilience.