IMU Preintegrated Features for Efficient Deep Inertial Odometry

Explore cutting-edge research on efficient deep inertial odometry in this 22-minute conference talk from the tinyML Research Symposium 2022. Delve into the innovative concept of IMU preintegrated features as presented by Rooholla Khorrambakht, a Robotics Researcher from K.N.Toosi University of Technology. Learn about measurement errors, problem statements, and the core concepts behind this technology. Discover the evaluation methods, training processes, and applications in human walking scenarios. Gain insights into the latest results and participate in a Q&A session to deepen your understanding of this advanced topic in robotics and machine learning.

Introduction
Measurement Errors
Problem Statement
Concept
Evaluation
Training
Human Walking
Results
Questions