STEP: State Estimator for Legged Robots Using a Preintegrated Foot Velocity Factor

Kim, Yeeun; Yu, Byeongho; Lee, Eungchang Mason; Kim, Joon-ha; Park, Hae Won; Myung, Hyun

doi:10.1109/lra.2022.3150844

Cited by 21 publications

(11 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two robots collected datasets in several outdoor environments while traveling over 1.5 km with an average velocity of 0.5 m/s. Note that our robots move at a much faster speed than prior works (for example, [9] is 0.125 m/s and [4] is 0.25 m/s). In each dataset, the robot moves in a large loop and we evaluate the final position estimation drift after the robot returns to the starting point.…”

Section: B Outdoor Experimentsmentioning

confidence: 99%

“…In each dataset, the robot moves in a large loop and we evaluate the final position estimation drift after the robot returns to the starting point. We also note that 1% drift is equivalent to 0.1m of the 10M Relative Translation Error (RTE) metric used in [4] and [9]. Details of datasets can be found in the open-source code base.…”

Section: B Outdoor Experimentsmentioning

confidence: 99%

“…Observation models combined with a dynamics model of the robot form a factor graph [6] describing a nonlinear optimization problem whose solution is the maximum-likelihood state estimate. Prior work [7]- [9] has shown that VILO outperforms methods that only utilize a subset of the aforementioned sensors, such as Visual-Inertial-Odometry (VIO) [10] or Leg Odometry (LO) [3] alone.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Cerberus: Low-Drift Visual-Inertial-Leg Odometry For Agile Locomotion

Yang¹,

Zhang²,

Fu³

et al. 2022

Preprint

View full text Add to dashboard Cite

We present an open-source Visual-Inertial-Leg Odometry (VILO) state estimation solution, Cerberus, for legged robots that estimates position precisely on various terrains in real time using a set of standard sensors, including stereo cameras, IMU, joint encoders, and contact sensors. In addition to estimating robot states, we also perform online kinematic parameter calibration and contact outlier rejection to substantially reduce position drift. Hardware experiments in various indoor and outdoor environments validate that calibrating kinematic parameters within the Cerberus can reduce estimation drift to lower than 1% during long distance high speed locomotion. Our drift results are better than any other state estimation method using the same set of sensors reported in the literature. Moreover, our state estimator performs well even when the robot is experiencing large impacts and camera occlusion. The implementation of the state estimator, along with the datasets used to compute our results, are available at https://github.com/ShuoYangRobotics/Cerberus.

show abstract

Section: B Outdoor Experimentsmentioning

confidence: 99%

Section: B Outdoor Experimentsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Cerberus: Low-Drift Visual-Inertial-Leg Odometry For Agile Locomotion

Yang¹,

Zhang²,

Fu³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…State estimation from only leg odometry and IMU such as in [1], [8], [9], [10] has limitations in observability of state variables such as yaw rotation or absolute position in a world reference frame. To this end, several approaches combine proprioceptive and IMU measurements with exteroceptive sensors such as vision [11], [12], [13], [14], [15], LiDAR [16], or both [4], [17]. Vision sensors are particularly lightweight compared to LiDARs.…”

Section: Related Workmentioning

confidence: 99%

“…The approach uses visual odometry estimates as relative pose factors. Kim et al [15] tightly integrate visual keypoint depth estimation with inertial measurement and preintegrated leg velocity factors. Our approach integrates absolute yaw and position measurements by the VIO, while height drift of the VIO wrt.…”

Section: Related Workmentioning

confidence: 99%

Visual-Inertial and Leg Odometry Fusion for Dynamic Locomotion

Victor¹,

Li²,

Khorshidi³

et al. 2022

Preprint

View full text Add to dashboard Cite

Implementing dynamic locomotion behaviors on legged robots requires a high-quality state estimation module. Especially when the motion includes flight phases, state-of-theart approaches fail to produce reliable estimation of the robot posture, in particular base height. In this paper, we propose a novel approach for combining visual-inertial odometry (VIO) with leg odometry in an extended Kalman filter (EKF) based state estimator. The VIO module uses a stereo camera and IMU to yield low-drift 3D position and yaw orientation and drift-free pitch and roll orientation of the robot base link in the inertial frame. However, these values have a considerable amount of latency due to image processing and optimization, while the rate of update is quite low which is not suitable for low-level control. To reduce the latency, we predict the VIO state estimate at the rate of the IMU measurements of the VIO sensor. The EKF module uses the base pose and linear velocity predicted by VIO, fuses them further with a second high-rate IMU and leg odometry measurements, and produces robot state estimates with a high frequency and small latency suitable for control. We integrate this lightweight estimation framework with a nonlinear model predictive controller and show successful implementation of a set of agile locomotion behaviors, including trotting and jumping at varying horizontal speeds, on a torque-controlled quadruped robot.

show abstract