A lightweight and scalable visual-inertial motion capture system using fiducial markers

He, Guiqin; Zhong, Shangkun; Guo, Jifeng

doi:10.1007/s10514-019-09834-7

Cited by 5 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since J is full-column rank, we can compute its pseudoinverse J + = (J J) −1 J to invert (24) to find Σ V in = J + Σ u J + . Then, assuming uncorrelated pixel noises, Σ u ∈ R 8×8 is a diagonal matrix with terms equal to σ 2 = n 2 , n being a number of pixels accounting for the pixelization noise and motion blur, as in [15] for instance. We obtain finally,…”

Section: B Factor Covariancementioning

confidence: 99%

“…In [23], the authors rely on a EKF in which state propagation in naturally handled by the IMU and each marker detection is used in an update step where the reprojection error of its 4 corners provides a 8D innovation vector. A closer solution to ours was very recently proposed in [15] and is also based on graph SLAM optimization benefiting from Forster's IMU pre-integration from GTSAM. As explained previously, the Apriltag factor formulation is different from ours and the algorithm is tested on large datasets consisting only of smooth motions.…”

Section: Related Workmentioning

confidence: 99%

“…While our final goal is to merge in the optimal estimator the measurements coming from all the sensors of the robot, we focus here on contributions validating the use of visual-inertial localization and mapping on a humanoid robot navigating indoor in a 3D environment. For the visual factor, we rely on April tags [14], [15], while proposing a practical contribution to avoid ambiguity issues in the pose estimation of the tags. For the inertial factor, we build upon Forster pre-integration [12] and propose an original and more rigorous theoretical formulation, by exhibiting a Lie topology that is suitable for optimal estimation.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Absolute humanoid localization and mapping based on IMU Lie group and fiducial markers

Fourmy¹,

Atchuthan²,

Mansard³

et al. 2019

2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids)

View full text Add to dashboard Cite

Current locomotion algorithms in structured (indoor) 3D environments require an accurate localization. The several and diverse sensors typically embedded on legged robots (IMU, coders, vision and/or LIDARS) should make it possible if properly fused. Yet this is a difficult task due to the heterogeneity of these sensors and the real-time requirement of the control. While previous works were using staggered approaches (odometry at high frequency, sparsely corrected from vision and LIDAR localization), the recent progress in optimal estimation, in particular in visual-inertial localization, is paving the way to a holistic fusion. This paper is a contribution in this direction. We propose to quantify how a visual-inertial navigation system can accurately localize a humanoid robot in a 3D indoor environment tagged with fiducial markers. We introduce a theoretical contribution strengthening the formulation of Forster's IMU pre-integration, a practical contribution to avoid possible ambiguity raised by pose estimation of fiducial markers, and an experimental contribution on a humanoid dataset with ground truth. Our system is able to localize the robot with less than 2 cm errors once the environment is properly mapped. This would naturally extend to additional measurements corresponding to leg odometry (kinematic factors) thanks to the genericity of the proposed pre-integration algebra.

show abstract

Section: B Factor Covariancementioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Absolute humanoid localization and mapping based on IMU Lie group and fiducial markers

Fourmy¹,

Atchuthan²,

Mansard³

et al. 2019

2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids)

View full text Add to dashboard Cite

show abstract

“…Nevertheless, the fact of being extremely low-cost, robust, fast, and easy-to-use makes AR markers an interesting tool that can be used for pose estimation in several applications. Currently, such systems are mostly used for AR [100] and robot localization [101,102]. Very few studies have explored the potential of using AR markers for the purpose of human motion analysis, which might be much further challenging [69,103,98].…”

Section: Fig 26 Aruco Markersmentioning

confidence: 99%

Toward an Affordable Multi-Modal Motion Capture System Framework for Human Kinematics and Kinetics Assessment

Mallat

Bonnet

Khalil

et al. 2018

Biosystems &Amp; Biorobotics

View full text Add to dashboard Cite

Quantifying human motor act starts with measuring and estimating kinematics and dynamics variables as accurately as possible. Monitoring human motion has a wide array of applications in functional rehabilitation, orthopaedics, sports, assistive robotics or industrial ergonomics. Today's motion capture systems usually refer to stereophotogrammetric systems and laboratory-grade force-plate that are accurate but also costly, require expert skills, and are not portable. Recently, the use of affordable sensors for human motion estimation, such as Inertial Measurement Unit or RGB-Depth camera(s), has been the subject of numerous studies. Despite their great potential to be used outside of the laboratory, these systems still suffer from limited accuracy, mainly due to inherent IMU drift and visual occlusions, and the joint kinematics and kinetics estimates are still difficult to be estimated. These drawbacks might explain why such systems are rarely used in common clinical applications or for in-home rehabilitation programs. In this context, this thesis deals with the development of a new affordable motion capture system capable of estimating accurately human 3D joint state. Unlike previous studies based on either visual or inertial sensors, the proposed approach consists in combining data from newly designed visual-inertial sensors. The system is also making use of new practical calibration methods, which do not require any external equipment while remaining very affordable. All sensors data are fused into a constrained extended Kalman filter that takes advantage of the biomechanics of the human body and of the investigated tasks to improve significantly joint state estimate. This is done by incorporating different types of constraints, such as joint limits, rigid-body and soft joint constraints, as well as modelling the temporal evolution of joint trajectories and/or sensors random bias. The system's ability to estimate accurate 3D joint kinematics has been validated through various case studies of daily life activities for upper-arm and treadmill gait. Two different prototypes with different sensors count and configurations have been investigated. Experiments conducted with several healthy subjects showed very satisfactory results when compared to a gold standard motion capture system. Overall, the average RMS difference between the two systems was below 4deg. This was also the case when a reduced number of sensors was used for gait analysis. This system was also used for the dynamics identification of a viii lower-limbs human-exoskeleton system. As a result, an error below 6% was observed when comparing estimated and measured external ground reaction force and moments. Finally, beyond these validations, a dynamics assessment framework has been proposed with the aim of selecting an optimal human-exoskeleton dynamic model that is the best trade-off between the accuracy of kinetic estimation, i.e., joint torque, and simplicity of modelling. To this end, the proposed framework consists in quantifying the independent con...

show abstract

“…Fiducial tags, or fiducial markers, are used in computer vision (CV) applications for robot localization [1,2], mapping and localization of large environments [3][4][5], or for pose estimation in medical endoscopy [6]. Markers are also used for metric purposes, e.g., for calibration [7] and monitoring changes in distances and orientations in historic structures [8].…”

Section: Introductionmentioning

confidence: 99%

Optimizing the Maximal Perturbation in Point Sets while Preserving the Order Type

Fraga

Hernández

2019

MCA

View full text Add to dashboard Cite

Recently a new kind of fiducial marker based on order type (OT) has been proposed. Using OT one can unequivocally identify a set of points through its triples of point orientation, and therefore, there is no need to use metric information. These proposed order type tags (OTTs) are invariant under a projective transformation which allows identification of them directly from a photograph. The magnitude of noise in the point positions that a set of points can support without changing its OT, is named the maximal perturbation (MP) value. This value represents the maximal displacement that any point in the set can have in any direction without changing the triplet’s orientation in the set. A higher value of the MP makes an OTT instance more robust to perturbations in the points positions. In this paper, we address the problem of how to improve the MP value for sets of points. We optimize “by hand” the MP for all the 16 subsets of points in the set of OTs composed of six points, and we also propose a general algorithm to optimize all the sets of OTs composed of six, seven, and eight points. Finally, we show several OTTs with improved MP values, and their use in an augmented reality application.

show abstract

A lightweight and scalable visual-inertial motion capture system using fiducial markers

Cited by 5 publications

References 19 publications

Absolute humanoid localization and mapping based on IMU Lie group and fiducial markers

Absolute humanoid localization and mapping based on IMU Lie group and fiducial markers

Toward an Affordable Multi-Modal Motion Capture System Framework for Human Kinematics and Kinetics Assessment

Optimizing the Maximal Perturbation in Point Sets while Preserving the Order Type

Contact Info

Product

Resources

About