1. 2019: The Year of Machine Learning?
We already announced that we are working on the first inertial sensor with Finite State Machines and a decision tree, which would effectively bring machine learning to MEMS for the first time. We also made a lot of headway into the implementations of specific neural network topologies onto embedded systems, as we saw with our SoC Orlando in 2018. Machine learning itself continues to be a significant topic of discussion because it opens the door to efficiencies and features that were just impossible until a few years ago. It will thus be interesting to see how this branch of artificial intelligence shapes tomorrow’s embedded systems.
- Machine Learning Inside an Inertial Sensor for Unbelievable Efficiency
- Orlando: An SoC With a Neural Network and the Future of IoT?