Want to accurately predict the remaining useful life (RUL) of lithium-ion batteries — in real-time and at the edge?
Forlinx Embedded has integrated a CNN + LSTM AI algorithm with our powerful RK3588-based FET3588-C SoM, enabling ultra-fast (0.55 ms/sample) and highly accurate RUL prediction.
✅ 6 TOPS NPU for edge AI inference
✅ MAPE as low as 3.3%
✅ FP16 quantized model optimized via RKNN
✅ Based on real NASA battery aging datasets
Whether you're building EV battery systems, smart energy platforms, or industrial electronics, this solution delivers robust AI performance and low power consumption.
More info:
https://www.forlinx.net/industrial-news/battery-rul-prediction-rk3588-702.html
Top comments (0)