FusionSense: Tri-Stage Near-Sensor Learning for Runtime-Adaptive Multimodal Edge Intelligence

Sanggeon Yun, Ryozo Masukawa, Minhyoung Na, Hyunwoo Oh, Yoshiki Yamaguchi, Wenjun Huang, SungHeon Jeong, Mohsen Imani

arXiv:2605.22868 · 2026-05-25 공개 · arXiv · PDF

edge-computing multimodal-fusion edge-intelligence runtime-adaptivity energy-constrained rgb-depth fusion-aware filter-out-safe

Abstract

Autonomous systems and smart-industry deployments increasingly split computation across near-sensor, edge, and cloud resources, where tight energy, latency, and reliability budgets demand run-time adaptivity. In practice, deciding what to compute and transmit at each point is pivotal; yet as multimodal sensor suites (cameras, LiDAR/depth, etc.) proliferate at the edge, most prior approaches either (i) fuse modalities on powerful servers or (ii) apply uni-modal near-sensor filters that ignore cross-modal dependencies, leading to redundant transmissions or missed events. We present FusionSense, a fusion-aware intelligent sensing framework for energy-constrained autonomous edge systems. Lightweight near-sensor classifiers are trained via a three-step procedure: (i) a server-side fusion model learns the downstream task, (ii) filter-out-safe (FoS) labels quantify each modality's necessity relative to the fused decision, and (iii) an edge-side fusion model is compacted by injecting near-sensor predictions as auxiliary signals. The result is a run-time decision layer that jointly reduces compute and communication while scaling linearly with sensor count. On a dual-modality (RGB+Depth/LiDAR) setup with SynDrone, FusionSense sustains task quality at substantially higher data-reduction rates than uni-modal filters and delivers large end-to-end gains: up to 33x lower energy at 1% FoI prevalence, 11x at 10%, a 92.3% reduction in quality loss at a fixed 30% data reduction, and roughly 1.5x higher energy savings than the best prior filtering baseline.

한국어 요약

📋 한 줄 요약

**[Near-Sensor Fusion]** FusionSense가 server fusion → Filter-out-Safe 라벨링 → edge fusion 컴팩션의 3단계 학습으로 RGB+Depth/LiDAR 통합 — SynDrone에서 1% FoI prevalence 시 최대 33× 에너지 절감.

🎯 핵심 기여도

💡 핵심 아이디어

멀티모달 sensor fusion의 효율은 server·edge·sensor를 분리한 3단계 학습 — server fusion → FoS 라벨링 → edge fusion 컴팩션 — 으로 달성되며, 이는 cross-modal 의존성을 보존하면서 compute·통신을 동시 줄이고 센서 수에 선형으로 확장된다.

🔬 기술적 접근법

📊 주요 결과

💭 의의 및 한계

**의의**: 멀티모달 edge intelligence의 효율 break-through, run-time adaptivity 제공, 센서 수 선형 확장성. **한계**: SynDrone 환경 평가 중심, RGB+Depth/LiDAR dual modality에 집중 — 3+ 모달리티 일반화 추가 검증, FoS 라벨 품질에 의존.

🚀 실용적 활용