2019 International Conference on Robotics and Automation (ICRA) 2019
DOI: 10.1109/icra.2019.8793598
|View full text |Cite
|
Sign up to set email alerts
|

Unsupervised Gait Phase Estimation for Humanoid Robot Walking

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
9
0

Year Published

2019
2019
2024
2024

Publication Types

Select...
4
3
1

Relationship

2
6

Authors

Journals

citations
Cited by 18 publications
(9 citation statements)
references
References 23 publications
0
9
0
Order By: Relevance
“…Subsequently, the desired step locations are fed to the walking engine that computes in real-time the walking pattern (Piperakis et al, 2014 ) and tracks that pattern using onboard proprioceptive sensing such as the IMU, joint encoder, and pressure measurements (Piperakis and Trahanias, 2016 ; Piperakis et al, 2018 ) and/or external odometry measurements (Piperakis et al, 2019a , b ) along with the current contact status (Piperakis et al, 2019c ), to achieve fast and dynamically stable locomotion. The latter is vital to the success of the task since the humanoid carries a significant weight (mounted LIDAR and bowl with items) and still manages stable omnidirectional walk.…”
Section: Resultsmentioning
confidence: 99%
“…Subsequently, the desired step locations are fed to the walking engine that computes in real-time the walking pattern (Piperakis et al, 2014 ) and tracks that pattern using onboard proprioceptive sensing such as the IMU, joint encoder, and pressure measurements (Piperakis and Trahanias, 2016 ; Piperakis et al, 2018 ) and/or external odometry measurements (Piperakis et al, 2019a , b ) along with the current contact status (Piperakis et al, 2019c ), to achieve fast and dynamically stable locomotion. The latter is vital to the success of the task since the humanoid carries a significant weight (mounted LIDAR and bowl with items) and still manages stable omnidirectional walk.…”
Section: Resultsmentioning
confidence: 99%
“…with n v b ∼ N (0, R n k ) be the normal zero mean kinematic velocity noise with covariance R n k . The above measurements do not accumulate leg odometry drift during the gait and are commonly not contaminated with outliers when accurate contact states are estimated [30], [31]. Thus, we distinguish them with the superscript n for nominal measurements that will be not examined for outliers.…”
Section: B Measurement Modelmentioning
confidence: 99%
“…To efficiently consider the robot's kinematics and contact effects we employ the State Estimation Robot Walking (SEROW) framework [47,49,76]. The latter fuses IMU, joint encoder, and Force/Torque (F/T) measurements to accurately estimate the following state vector x t : The kinematic information derived from SEROW are used to improve the performance of the localization and mapping processes in visual SLAM.…”
Section: Methods and Resultsmentioning
confidence: 99%
“…In all state-of-the-art approaches, the objective was to determine whether a specific foot is in contact or not. Recently, in [76] we raised a broader question: in which gait phase is the robot currently in? To this end, we proposed a holistic framework, based on unsupervised learning from proprioceptive sensing that accurately and efficiently addresses this problem.…”
Section: Contact Detectionmentioning
confidence: 99%
See 1 more Smart Citation