FT-WBC: Learning Fault-Tolerant Whole-Body Control for Legged Loco-Manipulation

๐Ÿ“… 2026-06-23
๐Ÿ“ˆ Citations: 0
โœจ Influential: 0
๐Ÿ“„ PDF
๐Ÿค– AI Summary
Legged manipulation robots are prone to instability under actuator failures due to disturbances from the manipulator arm, and existing fault-tolerant control approaches struggle to simultaneously maintain base stability and preserve the armโ€™s workspace. This work proposes the FT-WBC framework, which introduces, for the first time, a fault-aware posture adaptation mechanism. By estimating faulty joints through proprioceptive history and leveraging a decoupled upperโ€“lower body policy architecture, the method translates destabilizing base posture requests into safe commands and generates compensatory gaits. Evaluated in both simulation and real-world environments, the approach significantly improves survival rates and arm reachability under faults, and demonstrates zero-shot transfer to physical robots without fine-tuning.
๐Ÿ“ Abstract
Legged manipulators combine the mobility of legged platforms with the manipulation capability of robotic arms. However, arm-induced Center-of-Mass shifts and dynamic disturbances make the system more prone to instability under actuator failures, potentially leading to falls, task failures, or safety risks. Existing fault-tolerant control methods mainly focus on locomotion alone, leaving the coupled problem of whole-body stability and arm reachability in fault-tolerant loco-manipulation largely unaddressed. To bridge this gap, we propose FT-WBC, a fault-tolerant loco-manipulation framework for robust whole-body control of legged manipulators under actuator failures. FT-WBC adopts a decoupled upper- and lower-body policy architecture and introduces two key modules: a Fault Estimator (FE) and a Posture Adaptation Module (PAM). The FE predicts faulty joints from lower-body proprioceptive histories, while the PAM uses this fault information to adapt the base posture plan generated by the arm policy, converting potentially unstable posture requests into safe and executable base posture commands. Through this fault-aware posture adaptation mechanism, FT-WBC synthesizes compensatory gaits under actuator failures and preserves as much arm workspace as possible while maintaining whole-body stability. Simulation and real-world experiments show that FT-WBC significantly improves survival rate and workspace under weakening or locked failures, and transfers zero-shot to a real legged manipulator in the real world.
Problem

Research questions and friction points this paper is trying to address.

fault-tolerant control
legged loco-manipulation
whole-body stability
actuator failures
arm reachability
Innovation

Methods, ideas, or system contributions that make the work stand out.

fault-tolerant control
whole-body control
legged loco-manipulation
posture adaptation
zero-shot transfer
๐Ÿ”Ž Similar Papers
No similar papers found.
Y
Yudong Zhong
The Hong Kong University of Science and Technology (Guangzhou)
P
Pengfei Mai
The Hong Kong University of Science and Technology (Guangzhou)
S
Sikai Guo
The Hong Kong University of Science and Technology (Guangzhou)
Jiahang Cao
Jiahang Cao
The University of Hong Kong
Robot LearningGenerative ModelsCognitive-inspired Models
Zhihai Bi
Zhihai Bi
Fudan University; HKUST(GZ)
RoboticsLoco-ManipulationReinforcement Learning
Q
Qiuyue Liu
The Hong Kong University of Science and Technology (Guangzhou)
Z
Ziyan Feng
The Hong Kong University of Science and Technology (Guangzhou)
Jinni Zhou
Jinni Zhou
HKUST(GZ), HKUST
Jun Ma
Jun Ma
Assistant Professor, The Hong Kong University of Science and Technology
RoboticsAutonomous DrivingMotion Planning and ControlOptimization