Neuromorphic computing specialist BrainChip has launched a reference platform for battery-powered industrial sensors and wearables. The AkidaTag platform uses neuromorphic AI to enable always-on sensing and on-device learning without requiring a cloud connection.

The system is designed for remote industrial environments where sensors must interpret data locally to preserve battery life and privacy. By using event-based digital processing, the platform can detect anomalies in vibration and motion for predictive maintenance on industrial equipment. It also supports voice-activated interfaces and acoustic environment classification.

“This platform proves that high-performance AI and long battery life are no longer mutually exclusive,” said Sean Hehir, CEO of BrainChip. He stated that the blueprint allows manufacturers to build devices that are always-on and capable of self-learning at the edge.

The hardware utilises the BrainChip AKD1500 neuromorphic co-processor alongside a Nordic Semiconductor nRF5340 wireless system-on-chip. This configuration handles connectivity and sensor management while maintaining extreme energy efficiency. The reference design includes hardware schematics, firmware, and a mobile application for model updates and diagnostics.

BrainChip stated that the platform addresses the limitations of “one-size-fits-all” AI models by allowing for on-device adaptive learning. The reference platform will be available for evaluation in May 2026, with volume production expected in the third quarter of the year.