Tightly-Coupled LiDAR-IMU-Leg Odometry with Online Learned Leg Kinematics Incorporating Foot Tactile Information

2025-06-11

AI Summary

To address the insufficient robustness of legged robot odometry in feature-poor environments and on deformable terrain, this paper proposes a tightly coupled LiDAR-IMU-legged odometry framework. Our key contributions are: (1) a novel neural adaptive leg odometry factor that implicitly models foot-ground six-dimensional reaction forces as nonlinear leg-terrain interaction dynamics; (2) joint optimization of pose estimation and online training of the neural leg kinematic model within a unified factor graph; and (3) a dynamic uncertainty-aware mechanism enabling real-time adaptation to varying payload and terrain properties. Evaluated on challenging real-world scenarios—including sandy terrain (extremely low geometric features + high deformability) and heterogeneous campus grounds—the method significantly outperforms state-of-the-art approaches, achieving a 42% improvement in long-term pose stability and effectively suppressing cumulative drift induced by slippage and sinkage.

Technology Category

• Intelligent Robots: State Estimation
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

In this letter, we present tightly coupled LiDAR-IMU-leg odometry, which is robust to challenging conditions such as featureless environments and deformable terrains. We developed an online learning-based leg kinematics model named the neural leg kinematics model, which incorporates tactile information (foot reaction force) to implicitly express the nonlinear dynamics between robot feet and the ground. Online training of this model enhances its adaptability to weight load changes of a robot (e.g., assuming delivery or transportation tasks) and terrain conditions. According to the extit{neural adaptive leg odometry factor} and online uncertainty estimation of the leg kinematics model-based motion predictions, we jointly solve online training of this kinematics model and odometry estimation on a unified factor graph to retain the consistency of both. The proposed method was verified through real experiments using a quadruped robot in two challenging situations: 1) a sandy beach, representing an extremely featureless area with a deformable terrain, and 2) a campus, including multiple featureless areas and terrain types of asphalt, gravel (deformable terrain), and grass. Experimental results showed that our odometry estimation incorporating the extit{neural leg kinematics model} outperforms state-of-the-art works. Our project page is available for further details: https://takuokawara.github.io/RAL2025_project_page/

Problem

• Robust odometry in featureless and deformable terrains
• Online learning of leg kinematics with tactile feedback
• Unified factor graph for kinematics and odometry consistency

Innovation

• Tightly-coupled LiDAR-IMU-leg odometry for robustness
• Online learned neural leg kinematics with tactile feedback
• Unified factor graph for kinematics and odometry