Field Robotics Center Seminar / Robotics Masters Speaking Qualifier

  • Gates Hillman Centers
  • Mauldin Auditorium 1305
  • Masters Student
  • Robotics Institute
  • Carnegie Mellon University
Master's Thesis Presentation

Toward Invariant Visual Inertial State Estimation using Information Sparsification

In this work, we address two current challenges in real-time visual-inertial odometry (VIO) systems - efficiency and accuracy. To this end, we present a novel approach to tightly couple visual and inertial measurements in a fixed-lag VIO framework using information sparsification. To bound computational complexity, fixed-lag smoothers perform marginalization of variables but consequently deteriorate accuracy and especially efficiency. Current state-of-the-art approaches work around this by selectively discarding measurements and marginalizing additional variables. However, such strategies are sub-optimal from an information-theoretic perspective. In contrast, our approach formulates an optimization based on Kullback-Leibler divergence to preserve most of the information. To validate our approach, we conduct extensive real-time drone tests and perform comparisons to current state-of-the-art fixed-lag VIO methods in the EuRoC visual-inertial dataset. The experimental results show that the proposed method achieves competitive and superior accuracy in almost all trials.

In achieving a more efficient and accurate state estimator, the second part of the work presents the on-going progress in formulating an optimization-based VIO system using Matrix Lie Groups. Inspired by the recently developed Invariant-EKF framework, the proposed framework presents better convergence and addresses the consistency problem commonly seen in EKF-based and fixed-lag frameworks. In particular, we provide detailed derivations of a novel IMU preintegration framework using the group affine properties. Simulation results show our proposed formulation allows the nonlinear optimizer to converge with significantly fewer iterations, as compared to the state-of-the-art IMU preintegration scheme.

Shih-Chieh (Jerry) Hsiung is an M.S. student in the Robot Perception Lab at the Robotics Institute advised by Prof. Michael Kaess. He is broadly interested in the intersection of information theory, probabilistic theory and nonlinear optimization. His current research is focused on improving optimization methods in visual-inertial state estimation. Prior to CMU, Jerry received his B.S. in Computer Science from Harvey Mudd College. 

For More Information, Please Contact: