Update Contents

[swapnilsayansaha]

Paper: https://www.researchgate.net/publication/360075622_TinyOdom_Hardware-Aware_Efficient_Neural_Inertial_Navigation?fbclid=IwAR3F5LhoDiXD6tDhyE2PLFDB1hgy0IBM6V5YIUFwva7TvUvHYDi7C0ryTB8

Code: https://github.com/nesl/tinyodom

In this paper, we take #tinyml to its limits, attempting to localize humans, vehicles, robots, drones, animals, etc. using nothing but inertial sensor data (inertial odometry) on-board low-end IoT devices. We fuse state-of-the-art #bayesian neural architecture search and temporal convolutional #neuralnetwork with a physics-centric sequence learning formulation to provide neural inertial odometry models that are 31-134x smaller, yet have 1.15x higher resolution!