YZYuhao Zhang

Yuhao Zhang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Yuhao Zhang has a total of 40 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Contrastive Feature Reasoning for EEG Classification in Asynchronous BCI-Controlled Humanoid Robot

    by Tao Ding, Yuhao Zhang, Xingwei Zhao, Jian Shi, Bo Tao, Yang Shi, Han Ding

    Asynchronous brain–computer interfaces (aBCIs) provide a direct neural interface for decoding human intention, enabling practical brain-controlled interaction with humanoid robots. The key challenge, in aBCI-controlled robotics without explicit triggers, lies in the robust perception and accurate discrimination of control states from nonstationary electroencephalography (EEG) with fuzzy temporal boundaries. This study proposes a contrastive feature reasoning (CFR) framework for online EEG classification, which leverages a dual contrastive learning strategy to hierarchically model the spatio-spectral representation and temporal reasoning of EEG states. Specifically, template-EEG pairs are constructed to train a contrastive feature extractor via self-supervised contrastive learning, yielding cross-subject robust perception of spatio-spectral embedding features. Building on these embedding features, a contrastive feature reasoner is designed to perform temporal reasoning by evaluating contrastive similarity between current and historical features, coupled with relative-threshold discrimination for EEG state classification. Experiments on self-acquired EEG data from 12 participants, along with a humanoid robot control task involving six subjects, demonstrate an average cross-subject online control success rate of 97.33%, validating the effectiveness of CFR as a pivotal EEG classification component for aBCI-driven humanoid interaction.

  • Model Predictive Control for Tracking Bounded References With Arbitrary Dynamics

    by Shibo Han, Bonan Hou, Yuhao Zhang, Xiaotong Shi, Xingwei Zhao

    In this article, a model predictive control (MPC) method is proposed for constrained linear systems to track bounded references generated by linear exosystems with arbitrary Lyapunov stable dynamics. In the proposed method, sudden changes in reference are taken into account and are addressed by introducing an artificial reference as an additional decision variable. The cost function penalizes both the artificial state error and the reference error, while the terminal constraint is imposed on the artificial state error and the artificial reference. We specify the requirements for the terminal constraint and the cost function to guarantee recursive feasibility of the proposed method and asymptotic stability of the tracking error. Then, periodic and non-periodic references are analyzed, and methods are developed to determine the required cost function and terminal constraint. Finally, the efficiency of the proposed MPC controller is demonstrated with simulation examples.

  • A High-Precision and Compliant Interaction Method for Robot Based on Model Predictive Impedance Control

    by Yuhao Zhang, Zhenwei Zhang, Licheng Hou, Yang Xiang, Qingmiao Zhu, Xingwei Zhao, Bo Tao

  • Deadlock-Aware Control for Multirobot Coordination With Multiple Safety Constraints

    by Zhenwei Zhang, Yuhao Zhang, Xingwei Zhao, Bo Tao, Han Ding

    Multi-robot coordination in shared workspaces is prone to deadlocks, which can compromise operational capabilities and task efficiency. Accurately determining the timing and spatial locations of deadlocks is essential for effective resolution, yet remains challenging due to dynamic robot interactions and growing system complexity. To this end, a distributed deadlockaware control framework is proposed for robots to detect and avoid deadlocks while maintaining safe task execution. First, deadlocks are characterized by analyzing undesired equilibria in robot dynamics under safety constraints imposed by multiple stacked control barrier functions (CBFs). Our analysis reveals two critical properties: 1) Deadlocks occur at intersections of all active CBF boundaries, and 2) Deadlocks arise when robot stabilizing force are confined within the conical hull formed by active safety forces. These theoretical insights underpin a new detection method that identifies potential deadlocks from conflicts between safety requirements and task objectives. Furthermore, a reactive deadlock avoidance method is designed to help robots escape and prevent entry into potential deadlock regions by adaptively modulating the stabilizing force. A generalized workflow is established to systematically address deadlocks across various multi-robot tasks. Simulation and hardware experiments are conducted on robots collaborating in dense environments to validate the framework's effectiveness in preventing task failures caused by deadlocks.


  • A Sensing Strategy Combining Kinematic Model With FBG Sensors for Continuum Robots

    by Licheng Hou, Xingwei Zhao, Teru Chen, Yuhao Zhang, Qing Ling, Bo Tao

    Continuum manipulators (CMs) are commonly used in surgical robots due to their compact size and flexibility. However, these features also lead to low rigidity and limited accuracy. To enhance localization accuracy, this paper proposes a particle filter-based localization algorithm that combines kinematic and fiber Bragg grating (FBG) sensors data. Firstly, a tiny (Φ 4 mm) CM embedded with FBG sensors was designed, its kinematic model was derived based on the constant curvature model, and its measurement results were obtained by calculating curvature, interpolation, and integration according to the FBG sensor data. Subsequently, an adaptive weighted particle filter algorithm is proposed, which can prevent erroneous outputs when the kinematic model fails A series of experiments were conducted to demonstrate the proposed method. Bending experiments showed that, when the model is valid, the proposed method reduces error by 50% compared to measurement alone, with an error consistently under 0.8 mm across experimental conditions regardless of model validity. In trajectory ablation experiments on phantoms, the proposed method aligned more accurately with the ablation path.


  • LESO-Based NMPC Tracking Control of Climbing Robot on Large Components With Variable Curvature

    by Ke Tan, Zeyu Gong, Bo Tao, Yuhao Zhang, Ying Shi, Zhenfeng Gu, Chong Wu, Han Ding

    Wheeled climbing robots have great application prospects in the machining of large components with variable curvature. However, its accurate motion control on variable curvature surfaces faces two fatal challenges. The varied contact states between the robot’s wheels and the variable curvature surfaces make it difficult to establish an accurate kinematics model. Additionally, there exists a different degree of robot slippage when the robot moves in different attitudes due to the dragging effect of gravity. To overcome the above problems, we first present a kinematics modeling method with instantaneous plane constraints on a variable curvature surface. Subsequently, a linear extended state observer (LESO)-based nonlinear model predictive control (NMPC) scheme is designed, in which the NMPC is used to calculate the nominal control inputs and the LESO is used to estimate and compensate for lumped disturbance brought by the robot slippage and surface constraints. Experiments on a real wind turbine blade with variable curvature show that the proposed control scheme can well eliminate the influence of lumped disturbance, and the climbing robots can achieve unbiased (AVG < 0.1 mm) and high-precision (RMSE < 2 mm) trajectory tracking. provides a solution for automatic and high-precision trajectory tracking for climbing robot with localization system in factories, offering the potential for high-quality machining of large components.


  • A Differential Geometry Modeling Method for Wheeled Climbing Robots Tracking Control on Variable Curvature Surfaces

    by Ke Tan, Zeyu Gong, Bo Tao, Yuhao Zhang, Xiaoshun Liu, Ying Shi, Zhenfeng Gu, Han Ding

    Wheeled climbing robots equipped with tools can achieve high-coverage machining of large components with free-form surfaces. Trajectory tracking is the foundation for machining. However, there is a lack of a universal yet accurate model for robots while moving on variable curvature surfaces, making it difficult to guarantee tracking performance. Thus, this article proposes a universal contact frame based on differential geometry and establishes a kinematics model that uses surface parameters as state variables for robots moving on variable curvature surfaces. Based on this model, a tracking controller utilizing surface parameters error is then designed, enabling it to be applicable for tracking of climbing robots on free-form surfaces. Experiments with a wheeled climbing robot on a portion of a real wind turbine blade with a radius of more than 1.1 m demonstrate the effectiveness of the proposed method, with a root mean squared error (RMSE) of tracking accuracy of less than 1 mm.


  • Ultrasound-Guided Robotic Autonomous Operation Based on Real-Time Deformation Tracking and Prediction

    by Teru Chen, Xingwei Zhao, Yuhao Zhang, Guo Zheng, Licheng Hou, Qing Ling, Bo Tao, Zhouping Yin

    Robotic ultrasound (US)-guided surgery is one of the most common clinical workflows for many diseases. Intraoperative tissue deformation is a key challenge faced by robots in realizing autonomous surgery. This study develops a US-guided robotic autonomous operation scheme based on deformation tracking and prediction, which can overcome the adverse effects of intraoperative deformation. First, a deformation tracking method based on real-time nonrigid image registration is proposed to track tissue and path deformation in real time. Simultaneously, we propose a deformation prediction model that is updated by the forgetting factor recursive least squares for predicting the next deformation according to historical tracked data. Then, a time-varying vector field navigation law is incorporated to guide the robot following the dynamically deformed surgical path, therefore compensating for deformation. To verify the performance of the proposed method, a dual-arm robotic US surgical navigation system prototype is constructed, and in vitro experiments in a deformable environment are conducted. The results demonstrated that our method can guide the robot to compensate for tissue deformation in real-time with a dynamic path following error within 1 mm.


  • Physical Human–Robot Interaction Based on Adaptive Impedance Control for Robotic-Assisted Total Hip Arthroplasty

    by Yiming Chen, Yuhao Zhang, Xingwei Zhao, Qiang Xie, Kun Yang, Bo Tao, Han Ding

    The performance of human–robot interaction can be enhanced by imposing force constraints through virtual fixture (VF) algorithms. In this article, an adaptive impedance-based VF scheme is proposed, aiming to improve the accuracy of human–robot interaction in robot-assisted total hip arthroplasty (RA-THA). We developed a multimode impedance controller that realizes four distinct functional modes according to boundary divisions of the VF: free dragging, position constraints, restriction, and restoration of manipulability. A variable impedance control algorithm that adaptively adjusts the damping, stiffness, and force limitations is proposed to achieve precise stopping of the robot at the VF boundary. The stability of the closed-loop system is proved via Lyapunov theory. Experimental results demonstrate that the controller delivers precise, smooth, and stable VF constraints, even at varying velocities and in complex force interactions. The VF constraints are accurate within 1 mm, effectively improving the accuracy of human–robot interaction in RA-THA.