Imagine a future where small robotic teams roam our infrastructure—bridges, pipelines, buildings, and satellites—detecting problems promptly, such as leaks and cracks. Teams of robots offer many advantages for inspection, including high parallelization, resilience to failure, and scalability. However, for effective inspection, robots need to know where they are, a problem known as localization. Accurate robot localization demands substantial computational resources. If most of an agent’s computation is used for localization, little remains for productive inspection tasks like finding cracks, measuring vibration, and monitoring rust accumulation. Inspired by biological systems, we introduce a novel cooperative localization mechanism that minimizes collective computation expenditure through self-organized sacrifice. Here, a few agents bear the computational burden of localization; through local interactions, they improve the inspection productivity of the swarm. Our approach adaptively maximizes inspection productivity for unconstrained trajectories in various dynamic interaction and environmental settings. We demonstrate optimality and robustness using mean-field analytical models and hardware experiments with metal climbing robots inspecting a 3D cylinder.