Hybrid Contact Preintegration for Visual-Inertial-Contact State Estimation within Factor Graphs

by   Ross Hartley, et al.

The factor graph framework is a convenient modeling technique for robotic state estimation and sensor fusion where states are represented as nodes and measurements are modeled as factors. In designing a sensor fusion framework using factor graphs for legged robots, one often has access to visual, inertial, encoders, and contact sensors. While visual-inertial odometry has been studied extensively in this framework, the addition of a preintegrated contact factor for legged robots has been proposed recently. In this work, to cope with the problem of switching contact frames which was not addressed previously, we propose a hybrid contact preintegration that does not require the addition of frequently broken contact factors into the estimation factor graph. This paper presents a novel method for preintegrating contact information though an arbitrary number of contact switches. The proposed hybrid modeling approach reduces the number of required variables in the nonlinear optimization problem by only requiring new states to be added alongside camera or selected keyframes. This method is evaluated using real experimental data collected from a Cassie-series robot where the trajectory of the robot produced by a motion capture system is used as a proxy for ground truth data. The evaluation shows that inclusion of the proposed preintegrated hybrid contact factor alongside visual-inertial navigation systems improves robustness to vision failure as well as the estimation accuracy for legged robots while its generalization makes it more accessible for legged platforms.


Legged Robot State-Estimation Through Combined Forward Kinematic and Preintegrated Contact Factors

State-of-the-art robotic perception systems have achieved sufficiently g...

Probabilistic Contact State Estimation for Legged Robots using Inertial Information

Legged robot navigation in unstructured and slippery terrains depends he...

PIVO: Probabilistic Inertial-Visual Odometry for Occlusion-Robust Navigation

This paper presents a novel method for visual-inertial odometry. The met...

Fast Active Monocular Distance Estimation from Time-to-Contact

Distance estimation is fundamental for a variety of robotic applications...

Robust Legged Robot State Estimation Using Factor Graph Optimization

Legged robots, specifically quadrupeds, are becoming increasingly attrac...

Deep Multi-Modal Contact Estimation for Invariant Observer Design on Quadruped Robots

This work reports on developing a deep learning-based contact estimator ...

Model Free Calibration of Wheeled Robots Using Gaussian Process

Robotic calibration allows for the fusion of data from multiple sensors ...

Please sign up or login with your details

Forgot password? Click here to reset