CAHEROS: Constraint-Aware HEuristic Approach for RObust Sensor Placement
Onat Güngör, Tajana Rosing, Barış Akşanlı · 2021 IEEE Sensors · 2021
Coverage-only sensor placement lacks two important aspects: uncertainty in sensor measurements and possible sensor break-down. To incorporate these components into a sensor placement framework, we first propose our robust sensor placement model. In order to solve this model for large instances efficiently, we adapt different heuristics where we integrate constraints into the heuristic framework. To the best of our knowledge, we are the first ones adapting constraint-aware heuristics for robust sensor placement. Experimentally, by using a factory shop floor, we show that the greedy algorithm outperforms other heuristics while requiring more computational overhead. Compared to the coverage-only approach, our robust heuristics improves system robustness by up to 48%. Our heuristics can find a solution 400x faster than conventional solvers while only being 5.5% less optimal.