Qi Xue
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Qi Xue has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Chen Wang
School of Cybersecurity, Faculty of Computing, Harbin Institute of Technology, Harbin 150001, China
3 shared publications - Qi Han
Harbin Institute of Technology
3 shared publications - You Li
School of Cybersecurity, Faculty of Computing, Harbin Institute of Technology, Harbin 150001, China
3 shared publications - Zhaohai Meng
No. 707 Research Institute, China Shipbuilding Industry Corporation, Tianjin, China
2 shared publications - Yichen Dai
Faculty of Computing, School of Cyberspace Science, Harbin Institute of Technology, Harbin, China
1 shared publication
Latest Publications
- Open Access
MagGAT: Structure-Aware Spatiotemporal Graph Attention for Motor-Induced Magnetic Interference Compensation in Quadruped Robots
Permanent-magnet synchronous motors enable precise motion control in compact robots, but their leakage fields can corrupt onboard magnetometer measurements. Physics-based compensation is effective under fixed geometric conditions, whereas quadruped locomotion continuously changes the motor–sensor relationship. We propose MagGAT, a structure-aware spatiotemporal graph attention network that estimates compensated magnetic field signals from motor states and near-field measurements. A boom-mounted far-field magnetometer provides a pseudo-reference for supervised training and field-level evaluation. MagGAT combines a global–local dual-branch graph attention module for multi-motor spatial coupling with a gated recurrent unit for temporal disturbance modeling. Experiments on a Unitree Go2 quadruped used synchronized near-field and far-field measurements, joint angles, angular velocities, and commanded torques collected in indoor and outdoor environments. Across three runs with different random seeds, MagGAT achieved mean residual three-axis magnetic field RMSE values of 87.72–89.41 nT across the three test sequences and outperformed a physics-based fixed-parameter model, a static neural network method, and a temporal convolutional baseline. Ablation studies confirmed that both spatial graph modeling and temporal modeling contributed to compensation accuracy. These results support structure-aware spatiotemporal learning for suppressing motor-induced magnetometer interference on compact legged robots.
Open AccessMagGAT: Structure-Aware Spatiotemporal Graph Attention for Motor-Induced Magnetic Interference Compensation in Quadruped Robots
Permanent-magnet synchronous motors enable precise motion control in compact robots, but their leakage fields can corrupt onboard magnetometer measurements. Physics-based compensation is effective under fixed geometric conditions, whereas quadruped locomotion continuously changes the motor–sensor relationship. We propose MagGAT, a structure-aware spatiotemporal graph attention network that estimates compensated magnetic field signals from motor states and near-field measurements. A boom-mounted far-field magnetometer provides a pseudo-reference for supervised training and field-level evaluation. MagGAT combines a global–local dual-branch graph attention module for multi-motor spatial coupling with a gated recurrent unit for temporal disturbance modeling. Experiments on a Unitree Go2 quadruped used synchronized near-field and far-field measurements, joint angles, angular velocities, and commanded torques collected in indoor and outdoor environments. Across three runs with different random seeds, MagGAT achieved mean residual three-axis magnetic field RMSE values of 87.72–89.41 nT across the three test sequences and outperformed a physics-based fixed-parameter model, a static neural network method, and a temporal convolutional baseline. Ablation studies confirmed that both spatial graph modeling and temporal modeling contributed to compensation accuracy. These results support structure-aware spatiotemporal learning for suppressing motor-induced magnetometer interference on compact legged robots.
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 23 (15455955) , 2788-2799
Magnetic Moment Encoding (MME): An Information-Augmented Magnetic Navigation Method
In automated guided vehicle (AGV) magnetic navigation systems, magnetic markers typically serve solely as spatial references, offering limited informational content. This paper presents a novel Magnetic Moment Encoding (MME) system, which enables magnetic markers to simultaneously function as both physical landmarks and information carriers. The system leverages three-dimensional magnetic moment parameters of permanent magnets to achieve high-capacity information encoding. Compared to conventional binary magnetic pole encoding methods, MME increases information density by approximately threefold. To address the challenge of decoding MME signals, we propose a neural network-based Magnetic Moment Decoder (MMD), which enables robust decoding of encoded information even in complex magnetic field environments. Integrated into AGV navigation, the MME system achieves a positioning accuracy of 11.88±9.12 mm and a path yaw error of 4.21 ± 2.23°, while also supporting real-time adaptive path planning in dynamic environments. This study introduces a new paradigm for high-density, low-power magnetic information interaction, with broad application potential in industrial automation and the Internet of Things.
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 23 (15455955) , 2788-2799
Magnetic Moment Encoding (MME): An Information-Augmented Magnetic Navigation Method
In automated guided vehicle (AGV) magnetic navigation systems, magnetic markers typically serve solely as spatial references, offering limited informational content. This paper presents a novel Magnetic Moment Encoding (MME) system, which enables magnetic markers to simultaneously function as both physical landmarks and information carriers. The system leverages three-dimensional magnetic moment parameters of permanent magnets to achieve high-capacity information encoding. Compared to conventional binary magnetic pole encoding methods, MME increases information density by approximately threefold. To address the challenge of decoding MME signals, we propose a neural network-based Magnetic Moment Decoder (MMD), which enables robust decoding of encoded information even in complex magnetic field environments. Integrated into AGV navigation, the MME system achieves a positioning accuracy of 11.88±9.12 mm and a path yaw error of 4.21 ± 2.23°, while also supporting real-time adaptive path planning in dynamic environments. This study introduces a new paradigm for high-density, low-power magnetic information interaction, with broad application potential in industrial automation and the Internet of Things.