Embodied Vision
Members
Publications
Solo12 demonstrating jumping and trotting gaits using visual-inertial state estimation [
] in two outdoor experiments. ©~IEEE. Reprinted, with permission, from [].
Visual-Inertial State Estimation for Legged Robots
] in two outdoor experiments. ©~IEEE. Reprinted, with permission, from [].
In our work [], we fuse high-rate pose (3D position and orientation) estimates of visual-inertial odometry (VIO, []) with velocity estimates from foot contacts and leg kinematics of a quadruped robot (Solo12). The kinematics allows for measuring the height above the ground, while the visual-inertial odometry complementarily estimates lateral motion and yaw rotation with respect to a local map.
], we fuse high-rate pose (3D position and orientation) estimates of visual-inertial odometry (VIO, []) with velocity estimates from foot contacts and leg kinematics of a quadruped robot (Solo12). The kinematics allows for measuring the height above the ground, while the visual-inertial odometry complementarily estimates lateral motion and yaw rotation with respect to a local map.
n/a
Members
Publications
Embodied Vision
Autonomous Motion
Movement Generation and Control
Conference Paper
Visual-Inertial and Leg Odometry Fusion for Dynamic Locomotion
Dhédin, V., Li, H., Khorshidi, S., Mack, L., Ravi, A. K. C., Meduri, A., Shah, P., Grimminger, F., Righetti, L., Khadiv, M., et al.
In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2023 (Published)
preprint
video
DOI
URL
BibTeX
Embodied Vision
Article
Visual-Inertial Mapping with Non-Linear Factor Recovery
Usenko, V., Demmel, N., Schubert, D., Stückler, J., Cremers, D.
IEEE Robotics and Automation Letters (RA-L), 5(2):422-429, 2020, presented at IEEE International Conference on Robotics and Automation (ICRA) 2020, preprint arXiv:1904.06504 (Published)
Code
Preprint
URL
BibTeX