# Learning Contact Representation for Leg Odometry

Source: [Learning Contact Representation for Leg Odometry](<https://www.sota2.com/research/paper/6a22589186173ad9c1468aa1>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-06T05:00:04.316Z

Authors: Emre Girgin, Cagri Kilic

Published: 2026-06-03T22:50:45.000Z

## Abstract

The estimation of odometry in legged robots depends on the assumption that the velocity of the foot with respect to the world remains zero during the stance phase. Feedback for the main body velocity is derived from the kinematic serial chain of the feet making accurate leg phase detection is a critical subproblem. A considerable number of studies employ ground reaction force sensors mounted at the tip of the foot to classify, yet these sensors may not be universally available for all legged robots. Additionally, these sensors are often unresponsive to unaccounted disturbances, such as slippage, while the foot remains in contact with the ground. In this study, we propose a self-supervised representation learning framework for contact detection that utilizes the standard sensor set of joint encoders without reliance on force sensor augmentations. We employ learned representations to model the stance and swing phases probabilistically. The experimental results obtained confirm the efficacy of the proposed self-supervised contact detector. Our framework exhibited superior performance in comparison to supervised methods which necessitate sensor set augmentation and labeling, as well as baseline probabilistic approaches. Additionally, we make our code available to the public.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Leg Odometry on Real World Dataset](<https://www.sota2.com/research/sota/leg-odometry-on-real-world-dataset.md>): Leg Odometry; Real-world dataset; ATE (m): 17.2.

- [Leg Odometry on TartanGround Simulation](<https://www.sota2.com/research/sota/leg-odometry-on-tartanground-simulation.md>): Leg Odometry; TartanGround Simulation; ATE (m): 4.09.

- [Contact Classification on TartanGround Sim (test)](<https://www.sota2.com/research/sota/contact-classification-on-tartanground-sim-test.md>): Contact Classification; TartanGround Sim (test); Precision: 98.5.

- [Contact Classification on Real-World (Physical)](<https://www.sota2.com/research/sota/contact-classification-on-real-world-physical.md>): Contact Classification; Real-World (Physical); Precision: 84.4.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2606.05501>)

[PDF](<https://arxiv.org/pdf/2606.05501.pdf>)
