Xiaonan Huang
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
Research Subjects
Xiaonan Huang has a total of 102 co-authors. Here are the top co-authors by number of shared publications:
- Carmel Majidi
Soft Machines Lab, Mechanical Engineering, Carnegie Mellon University, Pittsburgh, USA
7 shared publications - Mohammad Khalid Jawed
Department of Mechanical and Aerospace Engineering University of California Los Angeles California USA
4 shared publications - Rebecca Kramer-Bottiglio
Department of Mechanical Engineering and Materials Science, Yale University, New Haven, USA
4 shared publications - Xiaohao Xu
University of Michigan
4 shared publications - Joran W. Booth
Department of Mechanical Engineering and Materials Science, School of Engineering and Applied Science, Yale University, New Haven, CT, USA
3 shared publications Kitty Kumar
3 shared publications
Latest Publications
- 4 October 2026
- Published by Springer Nature
- p. 569-587
One Scene, Two Depths: Probing Geometric Ambiguity in Monocular Foundation Models
- 4 October 2026
- Published by Springer Nature
- p. 569-587
One Scene, Two Depths: Probing Geometric Ambiguity in Monocular Foundation Models
- Open Access
- 18 September 2026
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (38)
Geometry-controlled instability pathway selection in elastic helices enables fast, efficient robotic locomotion
Mechanical instabilities can convert slow actuation into rapid, amplified motion, but predicting whether a three-dimensional elastic structure buckles smoothly or snaps remains challenging. Here, we show that helix geometry and boundary loading jointly select continuous buckling or snap-through in elastic filaments. Combining nonlinear rod theory, simulations, and high-throughput robotic arm experiments, we establish a predictive pathway map and identify snap-through as a robust mechanism for repeatable motion amplification under simple boundary actuation. Guided by this map, we design a snap-actuated helical limb that converts small end rotations into rapid global reconfiguration and cyclic energy release. Integrated into an untethered robot, this limb enables sustained hopping across wood, cloth, acrylic, leather, grass, and sand, reaching 3.21 body lengths per second with an electrical cost of transport of 4.79. Controlled comparisons with a rigid-legged robot show that snap-through drives the gains in speed, efficiency, and terrain robustness. The same limb architecture also enables agile turning, repeated back-flips, and swimming.
- 18 September 2026
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (38)
Open AccessGeometry-controlled instability pathway selection in elastic helices enables fast, efficient robotic locomotion
Mechanical instabilities can convert slow actuation into rapid, amplified motion, but predicting whether a three-dimensional elastic structure buckles smoothly or snaps remains challenging. Here, we show that helix geometry and boundary loading jointly select continuous buckling or snap-through in elastic filaments. Combining nonlinear rod theory, simulations, and high-throughput robotic arm experiments, we establish a predictive pathway map and identify snap-through as a robust mechanism for repeatable motion amplification under simple boundary actuation. Guided by this map, we design a snap-actuated helical limb that converts small end rotations into rapid global reconfiguration and cyclic energy release. Integrated into an untethered robot, this limb enables sustained hopping across wood, cloth, acrylic, leather, grass, and sand, reaching 3.21 body lengths per second with an electrical cost of transport of 4.79. Controlled comparisons with a rigid-legged robot show that snap-through drives the gains in speed, efficiency, and terrain robustness. The same limb architecture also enables agile turning, repeated back-flips, and swimming.
- 10 August 2026
- Published by Springer Nature in Nature Machine Intelligence
- Vol. 8 (8) , 1290-1301
Impact-resistant, autonomous robots inspired by tensegrity architecture
- 10 August 2026
- Published by Springer Nature in Nature Machine Intelligence
- Vol. 8 (8) , 1290-1301
Impact-resistant, autonomous robots inspired by tensegrity architecture
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (7) , 8036-8043
Morphology-Aware Graph Reinforcement Learning for Tensegrity Robot Locomotion
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (7) , 8036-8043
Morphology-Aware Graph Reinforcement Learning for Tensegrity Robot Locomotion
- Open Access
Highly Deformable Proprioceptive Membrane for Real-Time 3D Shape Reconstruction
Reconstructing the three-dimensional (3D) geometry of object surfaces is essential for robot perception, yet vision-based approaches are generally unreliable under low illumination or occlusion. This limitation motivates the design of a proprioceptive membrane that conforms to the surface of interest and infers 3D geometry by reconstructing its own deformation. Conventional shape-aware membranes typically rely on resistive, capacitive, or magneto-sensitive mechanisms. However, these methods often encounter challenges such as structural complexity, limited compliance during large-scale deformation, and susceptibility to electromagnetic interference. This work presents a soft, flexible, and stretchable proprioceptive silicone membrane based on optical waveguide sensing. The membrane sensor integrates edge-mounted LEDs and centrally distributed photodiodes (PDs), interconnected via liquid-metal traces embedded within a multilayer elastomeric composite. Rich deformation-dependent light intensity signals are decoded by a data-driven model to recover the membrane geometry as a 3D point cloud. On a customized 140 mm square membrane, real-time reconstruction of large-scale out-of-plane deformation is achieved at 90 Hz with an average reconstruction error of 1.3 mm, measured by Chamfer distance, while maintaining accuracy for indentations up to 25 mm. The proposed framework provides a scalable, robust, and low-profile solution for global shape perception in deformable robotic systems.
Open AccessHighly Deformable Proprioceptive Membrane for Real-Time 3D Shape Reconstruction
Reconstructing the three-dimensional (3D) geometry of object surfaces is essential for robot perception, yet vision-based approaches are generally unreliable under low illumination or occlusion. This limitation motivates the design of a proprioceptive membrane that conforms to the surface of interest and infers 3D geometry by reconstructing its own deformation. Conventional shape-aware membranes typically rely on resistive, capacitive, or magneto-sensitive mechanisms. However, these methods often encounter challenges such as structural complexity, limited compliance during large-scale deformation, and susceptibility to electromagnetic interference. This work presents a soft, flexible, and stretchable proprioceptive silicone membrane based on optical waveguide sensing. The membrane sensor integrates edge-mounted LEDs and centrally distributed photodiodes (PDs), interconnected via liquid-metal traces embedded within a multilayer elastomeric composite. Rich deformation-dependent light intensity signals are decoded by a data-driven model to recover the membrane geometry as a 3D point cloud. On a customized 140 mm square membrane, real-time reconstruction of large-scale out-of-plane deformation is achieved at 90 Hz with an average reconstruction error of 1.3 mm, measured by Chamfer distance, while maintaining accuracy for indentations up to 25 mm. The proposed framework provides a scalable, robust, and low-profile solution for global shape perception in deformable robotic systems.
- Open Access
- 15 September 2025
- Published by Springer Nature in Research Square
SPPIRO: A Soft, Power-Autonomous, Proprioceptive In-Pipe RObot for Adaptive Inspection in Complex Pipeline Environments
Subterranean and industrial pipelines are critical infrastructure for transporting water, oil, and gas, yet their internal complexity—featuring varying diameters, sharp turns, and multi-directional junctions—poses significant challenges for conventional inspection robots. We present SPPIRO, a fully untethered soft robot designed for adaptive pipeline inspection in such environments. SPPIRO integrates compact, custom-designed PCBs for onboard power, sensing, and control, enabling fully untethered operation. Its anchoring modules use motor-driven six-bar linkages with 3D-printed, honeycomb-inspired soft wheels, allowing adaptation to pipeline diameters ranging from 140 mm to 200 mm. The propulsion module consists of three self-sensing pneumatic Kresling origami actuators arranged at 120◦, providing omnidirectional steering and real-time body shape estimation. Using a Piecewise Constant Curvature (PCC) model, SPPIRO reconstructs both its own configuration and the surrounding pipe geometry during motion. Liquid-metal-based capacitive sensors embedded in the feet enable contact-aware anchoring, even in opaque pipelines. Experiments demonstrate SPPIRO’s ability to traverse vertical and horizontal segments, 90-degree turns, and T-junctions, achieving speeds up to 7.86 mm/s—outperforming prior untethered soft in-pipe robots. These results position SPPIRO as a promising platform for autonomous, geometry-aware pipeline inspection.
- 15 September 2025
- Published by Springer Nature in Research Square
Open AccessSPPIRO: A Soft, Power-Autonomous, Proprioceptive In-Pipe RObot for Adaptive Inspection in Complex Pipeline Environments
Subterranean and industrial pipelines are critical infrastructure for transporting water, oil, and gas, yet their internal complexity—featuring varying diameters, sharp turns, and multi-directional junctions—poses significant challenges for conventional inspection robots. We present SPPIRO, a fully untethered soft robot designed for adaptive pipeline inspection in such environments. SPPIRO integrates compact, custom-designed PCBs for onboard power, sensing, and control, enabling fully untethered operation. Its anchoring modules use motor-driven six-bar linkages with 3D-printed, honeycomb-inspired soft wheels, allowing adaptation to pipeline diameters ranging from 140 mm to 200 mm. The propulsion module consists of three self-sensing pneumatic Kresling origami actuators arranged at 120◦, providing omnidirectional steering and real-time body shape estimation. Using a Piecewise Constant Curvature (PCC) model, SPPIRO reconstructs both its own configuration and the surrounding pipe geometry during motion. Liquid-metal-based capacitive sensors embedded in the feet enable contact-aware anchoring, even in opaque pipelines. Experiments demonstrate SPPIRO’s ability to traverse vertical and horizontal segments, 90-degree turns, and T-junctions, achieving speeds up to 7.86 mm/s—outperforming prior untethered soft in-pipe robots. These results position SPPIRO as a promising platform for autonomous, geometry-aware pipeline inspection.
- Open Access
- 14 September 2025
- Published by Wiley in Advanced Science
- Vol. 12 (41) , e10382
SPARC: A Soft, Proprioceptive, Agile Robot for 3D Climbing and Exploration with Precise Trajectory Following
Soft robots, like their natural counterparts, are well‐suited for exploring 3D terrains due to their large deformation, impact resistance, and adaptive adhesion, derived from their inherent compliance. However, developing soft robots that can navigate both horizontal and vertical surfaces with seamless transitions and precise trajectory tracking remains a challenge. SPARC, a Soft, Proprioceptive, Agile Robot is presented for 3D climbing and exploration with accurate trajectory following. SPARC integrates three parallel 3D‐printed Kresling origami actuators with suction cups for 3D actuation and adhesion across diverse surfaces. Leveraging the contraction–twisting coupling of Kresling structures, SPARC enables real‐time proprioception through inverse kinematics modeling. A dual closed‐loop control strategy combined with efficient online path planning allows precise pose control using onboard angle sensors and global position tracking. Experiments demonstrate SPARC's ability to follow curved and sharply angled trajectories with high precision'achieving 0.5% accuracy on horizontal and 3% on vertical surfaces'while carrying payloads exceeding twice its weight (500 g) on vertical climbs. Moreover, connecting two SPARC modules enables smooth ground‐to‐wall transitions. The seamless integration of large‐magnitude 3D actuation, proprioceptive state sensing, and efficient online path planning makes SPARC a unique soft robot capable of precise trajectory following and navigation in 3D terrains.
- 14 September 2025
- Published by Wiley in Advanced Science
- Vol. 12 (41) , e10382
Open AccessSPARC: A Soft, Proprioceptive, Agile Robot for 3D Climbing and Exploration with Precise Trajectory Following
Soft robots, like their natural counterparts, are well‐suited for exploring 3D terrains due to their large deformation, impact resistance, and adaptive adhesion, derived from their inherent compliance. However, developing soft robots that can navigate both horizontal and vertical surfaces with seamless transitions and precise trajectory tracking remains a challenge. SPARC, a Soft, Proprioceptive, Agile Robot is presented for 3D climbing and exploration with accurate trajectory following. SPARC integrates three parallel 3D‐printed Kresling origami actuators with suction cups for 3D actuation and adhesion across diverse surfaces. Leveraging the contraction–twisting coupling of Kresling structures, SPARC enables real‐time proprioception through inverse kinematics modeling. A dual closed‐loop control strategy combined with efficient online path planning allows precise pose control using onboard angle sensors and global position tracking. Experiments demonstrate SPARC's ability to follow curved and sharply angled trajectories with high precision'achieving 0.5% accuracy on horizontal and 3% on vertical surfaces'while carrying payloads exceeding twice its weight (500 g) on vertical climbs. Moreover, connecting two SPARC modules enables smooth ground‐to‐wall transitions. The seamless integration of large‐magnitude 3D actuation, proprioceptive state sensing, and efficient online path planning makes SPARC a unique soft robot capable of precise trajectory following and navigation in 3D terrains.
- 7 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 10 (9) , 8626-8633
Design of a Variable Stiffness Quasi-Direct Drive Cable-Actuated Tensegrity Robot
Tensegrity robots excel in tasks requiring extreme levels of deformability and robustness. However, there are challenges in state estimation and payload versatility due to their high number of degrees of freedom and unconventional shape. This paper introduces a modular three-bar tensegrity robot featuring a customizable payload design. Our tensegrity robot employs a novel Quasi-Direct Drive (QDD) cable actuator with low-stretch polymer cables to achieve accurate proprioception without needing external force or torque sensors. The design allows for on-the-fly stiffness tuning for better environment and payload adaptability. In this paper, we present the robot's design, fabrication, assembly, and experimental results. Experimental data demonstrates the high accuracy cable length estimation ($< $1% error relative to bar length) and variable stiffness control of the cable actuator up to 7 times the minimum stiffness for self support. The shape morphing and stiffness tuning capabilities are leveraged in two realistic demonstrations. The presented tensegrity robot is a platform for future advancements in autonomous operation and open-source module design. Open source design files are available at https://github.com/UMich-HDRLab/tensegrity-robot-hardware/.
- 7 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 10 (9) , 8626-8633
Design of a Variable Stiffness Quasi-Direct Drive Cable-Actuated Tensegrity Robot
Tensegrity robots excel in tasks requiring extreme levels of deformability and robustness. However, there are challenges in state estimation and payload versatility due to their high number of degrees of freedom and unconventional shape. This paper introduces a modular three-bar tensegrity robot featuring a customizable payload design. Our tensegrity robot employs a novel Quasi-Direct Drive (QDD) cable actuator with low-stretch polymer cables to achieve accurate proprioception without needing external force or torque sensors. The design allows for on-the-fly stiffness tuning for better environment and payload adaptability. In this paper, we present the robot's design, fabrication, assembly, and experimental results. Experimental data demonstrates the high accuracy cable length estimation ($< $1% error relative to bar length) and variable stiffness control of the cable actuator up to 7 times the minimum stiffness for self support. The shape morphing and stiffness tuning capabilities are leveraged in two realistic demonstrations. The presented tensegrity robot is a platform for future advancements in autonomous operation and open-source module design. Open source design files are available at https://github.com/UMich-HDRLab/tensegrity-robot-hardware/.
- Open Access
Visual Anomaly Detection under Complex View-Illumination Interplay: A Large-Scale Benchmark
The practical deployment of Visual Anomaly Detection (VAD) systems is hindered by their sensitivity to real-world imaging variations, particularly the complex interplay between viewpoint and illumination which drastically alters defect visibility. Current benchmarks largely overlook this critical challenge. We introduce Multi-View Multi-Illumination Anomaly Detection (M2AD), a new large-scale benchmark comprising 119,880 high-resolution images designed explicitly to probe VAD robustness under such interacting conditions. By systematically capturing 999 specimens across 10 categories using 12 synchronized views and 10 illumination settings (120 configurations total), M2AD enables rigorous evaluation. We establish two evaluation protocols: M2AD-Synergy tests the ability to fuse information across diverse configurations, and M2AD-Invariant measures single-image robustness against realistic view-illumination effects. Our extensive benchmarking shows that state-of-the-art VAD methods struggle significantly on M2AD, demonstrating the profound challenge posed by view-illumination interplay. This benchmark serves as an essential tool for developing and validating VAD methods capable of overcoming real-world complexities. Our full dataset and test suite will be released at https://hustcyq.github.io/M2AD to facilitate the field.
Open AccessVisual Anomaly Detection under Complex View-Illumination Interplay: A Large-Scale Benchmark
The practical deployment of Visual Anomaly Detection (VAD) systems is hindered by their sensitivity to real-world imaging variations, particularly the complex interplay between viewpoint and illumination which drastically alters defect visibility. Current benchmarks largely overlook this critical challenge. We introduce Multi-View Multi-Illumination Anomaly Detection (M2AD), a new large-scale benchmark comprising 119,880 high-resolution images designed explicitly to probe VAD robustness under such interacting conditions. By systematically capturing 999 specimens across 10 categories using 12 synchronized views and 10 illumination settings (120 configurations total), M2AD enables rigorous evaluation. We establish two evaluation protocols: M2AD-Synergy tests the ability to fuse information across diverse configurations, and M2AD-Invariant measures single-image robustness against realistic view-illumination effects. Our extensive benchmarking shows that state-of-the-art VAD methods struggle significantly on M2AD, demonstrating the profound challenge posed by view-illumination interplay. This benchmark serves as an essential tool for developing and validating VAD methods capable of overcoming real-world complexities. Our full dataset and test suite will be released at https://hustcyq.github.io/M2AD to facilitate the field.
- 5 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 10 (4) , 4069-4076
Tensegrity Robot Proprioceptive State Estimation With Geometric Constraints
Tensegrity robots,characterized by a synergistic assembly of rigid rods and elastic cables, form robust structures that are resistant to impacts. However, this design introduces complexities in kinematics and dynamics, complicating control and state estimation. This work presents a novel proprioceptive state estimator for tensegrity robots. The estimator initially uses the geometric constraints of 3-bar prism tensegrity structures, combined with IMU and motor encoder measurements, to reconstruct the robot's shape and orientation. It then employs a contact-aided invariant extended Kalman filter with forward kinematics to estimate the global position and orientation of the tensegrity robot. The state estimator's accuracy is assessed against ground truth data in both simulated environments and real-world tensegrity robot applications. It achieves an average drift percentage of 4.2%, comparable to the state estimation performance of traditional rigid robots. This state estimator advances the state-of-the-art in tensegrity robot state estimation and has the potential to run in real-time using onboard sensors, paving the way for full autonomy of tensegrity robots in unstructured environments.
- 5 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 10 (4) , 4069-4076
Tensegrity Robot Proprioceptive State Estimation With Geometric Constraints
Tensegrity robots,characterized by a synergistic assembly of rigid rods and elastic cables, form robust structures that are resistant to impacts. However, this design introduces complexities in kinematics and dynamics, complicating control and state estimation. This work presents a novel proprioceptive state estimator for tensegrity robots. The estimator initially uses the geometric constraints of 3-bar prism tensegrity structures, combined with IMU and motor encoder measurements, to reconstruct the robot's shape and orientation. It then employs a contact-aided invariant extended Kalman filter with forward kinematics to estimate the global position and orientation of the tensegrity robot. The state estimator's accuracy is assessed against ground truth data in both simulated environments and real-world tensegrity robot applications. It achieves an average drift percentage of 4.2%, comparable to the state estimation performance of traditional rigid robots. This state estimator advances the state-of-the-art in tensegrity robot state estimation and has the potential to run in real-time using onboard sensors, paving the way for full autonomy of tensegrity robots in unstructured environments.