Xiaobo Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Xiaobo Liu has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Chaoyang Song
Department of Robotics, Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi, United Arab Emirates
6 shared publications - Fang Wan
School of Design, Southern University of Science and Technology, Shenzhen, China
4 shared publications - Xudong Han
School of Design, Southern University of Science and Technology, Shenzhen, China
4 shared publications - Ning Guo
SUSTech Institute of Robotics Southern University of Science and Technology Shenzhen China
3 shared publications Fang Wan
Southern University of Science and Technology, Shenzhen, Guangdong, China
2 shared publications- Jian Lin
State Key Laboratory of Tropical Oceanography, South China Sea Institute of Oceanology, Chinese Academy of Sciences, Guangzhou, China
2 shared publications
Latest Publications
- Open Access
- 13 March 2024
- Published by SAGE Publications in The International Journal of Robotics Research
- Vol. 43 (12) , 1916-1935
Proprioceptive learning with soft polyhedral networks
Proprioception is the “sixth sense” that detects limb postures with motor neurons. It requires a natural integration between the musculoskeletal systems and sensory receptors, which is challenging among modern robots that aim for lightweight, adaptive, and sensitive designs at low costs in mechanical design and algorithmic computation. Here, we present the Soft Polyhedral Network with an embedded vision for physical interactions, capable of adaptive kinesthesia and viscoelastic proprioception by learning kinetic features. This design enables passive adaptations to omni-directional interactions, visually captured by a miniature high-speed motion-tracking system embedded inside for proprioceptive learning. The results show that the soft network can infer real-time 6D forces and torques with accuracies of 0.25/0.24/0.35 N and 0.025/0.034/0.006 Nm in dynamic interactions. We also incorporate viscoelasticity in proprioception during static adaptation by adding a creep and relaxation modifier to refine the predicted results. The proposed soft network combines simplicity in design, omni-adaptation, and proprioceptive sensing with high accuracy, making it a versatile solution for robotics at a low material cost with more than one million use cycles for tasks such as sensitive and competitive grasping and touch-based geometry reconstruction. This study offers new insights into vision-based proprioception for soft robots in adaptive grasping, soft manipulation, and human-robot interaction.
- 13 March 2024
- Published by SAGE Publications in The International Journal of Robotics Research
- Vol. 43 (12) , 1916-1935
Open AccessProprioceptive learning with soft polyhedral networks
Proprioception is the “sixth sense” that detects limb postures with motor neurons. It requires a natural integration between the musculoskeletal systems and sensory receptors, which is challenging among modern robots that aim for lightweight, adaptive, and sensitive designs at low costs in mechanical design and algorithmic computation. Here, we present the Soft Polyhedral Network with an embedded vision for physical interactions, capable of adaptive kinesthesia and viscoelastic proprioception by learning kinetic features. This design enables passive adaptations to omni-directional interactions, visually captured by a miniature high-speed motion-tracking system embedded inside for proprioceptive learning. The results show that the soft network can infer real-time 6D forces and torques with accuracies of 0.25/0.24/0.35 N and 0.025/0.034/0.006 Nm in dynamic interactions. We also incorporate viscoelasticity in proprioception during static adaptation by adding a creep and relaxation modifier to refine the predicted results. The proposed soft network combines simplicity in design, omni-adaptation, and proprioceptive sensing with high accuracy, making it a versatile solution for robotics at a low material cost with more than one million use cycles for tasks such as sensitive and competitive grasping and touch-based geometry reconstruction. This study offers new insights into vision-based proprioception for soft robots in adaptive grasping, soft manipulation, and human-robot interaction.
- Open Access
- 22 January 2024
- Published by Wiley in Advanced Intelligent Systems
- Vol. 6 (1)
Autoencoding a Soft Touch to Learn Grasping from On‐Land to Underwater
- 22 January 2024
- Published by Wiley in Advanced Intelligent Systems
- Vol. 6 (1)
Open AccessAutoencoding a Soft Touch to Learn Grasping from On‐Land to Underwater
- Open Access
- 11 January 2024
- Published by Elsevier in Materials & Design
- Vol. 238
Vision-based tactile intelligence with soft robotic metamaterial
- 11 January 2024
- Published by Elsevier in Materials & Design
- Vol. 238
Open AccessVision-based tactile intelligence with soft robotic metamaterial
- Open Access
- 22 October 2023
- Published by Wiley in Advanced Intelligent Systems
- Vol. 6 (1)
Autoencoding a Soft Touch to Learn Grasping from On‐Land to Underwater
Robots play a critical role as the physical agent of human operators in exploring the ocean. However, it remains challenging to grasp objects reliably while fully submerging under a highly pressurized aquatic environment with little visible light, mainly due to the fluidic interference on the tactile mechanics between the finger and object surfaces. This study investigates the transferability of grasping knowledge from on‐land to underwater via a vision‐based soft robotic finger that learns 6D forces and torques (FT) using a supervised variational autoencoder (SVAE). A high‐framerate camera captures the whole‐body deformations while a soft robotic finger interacts with physical objects on‐land and underwater. Results show that the trained SVAE model learns a series of latent representations of the soft mechanics transferable from land to water, presenting a superior adaptation to the changing environments against commercial FT sensors. Soft, delicate, and reactive grasping enabled by tactile intelligence enhances the gripper's underwater interaction with improved reliability and robustness at a much‐reduced cost, paving the path for learning‐based intelligent grasping to support fundamental scientific discoveries in environmental and ocean research.
- 22 October 2023
- Published by Wiley in Advanced Intelligent Systems
- Vol. 6 (1)
Open AccessAutoencoding a Soft Touch to Learn Grasping from On‐Land to Underwater
Robots play a critical role as the physical agent of human operators in exploring the ocean. However, it remains challenging to grasp objects reliably while fully submerging under a highly pressurized aquatic environment with little visible light, mainly due to the fluidic interference on the tactile mechanics between the finger and object surfaces. This study investigates the transferability of grasping knowledge from on‐land to underwater via a vision‐based soft robotic finger that learns 6D forces and torques (FT) using a supervised variational autoencoder (SVAE). A high‐framerate camera captures the whole‐body deformations while a soft robotic finger interacts with physical objects on‐land and underwater. Results show that the trained SVAE model learns a series of latent representations of the soft mechanics transferable from land to water, presenting a superior adaptation to the changing environments against commercial FT sensors. Soft, delicate, and reactive grasping enabled by tactile intelligence enhances the gripper's underwater interaction with improved reliability and robustness at a much‐reduced cost, paving the path for learning‐based intelligent grasping to support fundamental scientific discoveries in environmental and ocean research.
- Open Access
- 21 October 2023
- Published by MDPI AG in Biomimetics
- Vol. 8 (6) , 501
Bio-Inspired Proprioceptive Touch of a Soft Finger with Inner-Finger Kinesthetic Perception
In-hand object pose estimation is challenging for humans and robots due to occlusion caused by the hand and object. This paper proposes a soft finger that integrates inner vision with kinesthetic sensing to estimate object pose inspired by human fingers. The soft finger has a flexible skeleton and skin that adapts to different objects, and the skeleton deformations during interaction provide contact information obtained by the image from the inner camera. The proposed framework is an end-to-end method that uses raw images from soft fingers to estimate in-hand object pose. It consists of an encoder for kinesthetic information processing and an object pose and category estimator. The framework was tested on seven objects, achieving an impressive error of 2.02 mm and 11.34 degrees for pose error and 99.05% for classification.
- 21 October 2023
- Published by MDPI AG in Biomimetics
- Vol. 8 (6) , 501
Open AccessBio-Inspired Proprioceptive Touch of a Soft Finger with Inner-Finger Kinesthetic Perception
In-hand object pose estimation is challenging for humans and robots due to occlusion caused by the hand and object. This paper proposes a soft finger that integrates inner vision with kinesthetic sensing to estimate object pose inspired by human fingers. The soft finger has a flexible skeleton and skin that adapts to different objects, and the skeleton deformations during interaction provide contact information obtained by the image from the inner camera. The proposed framework is an end-to-end method that uses raw images from soft fingers to estimate in-hand object pose. It consists of an encoder for kinesthetic information processing and an object pose and category estimator. The framework was tested on seven objects, achieving an impressive error of 2.02 mm and 11.34 degrees for pose error and 99.05% for classification.
- Open Access
DeepClaw: A Robotic Hardware Benchmarking Platform for Learning Object Manipulation
We present DeepClaw as a reconfigurable benchmark of robotic hardware and task hierarchy for robot learning. The DeepClaw benchmark aims at a mechatronics perspective of the robot learning problem, which features a minimum design of robot cell that can be easily reconfigured to host robot hardware from various vendors, including manipulators, grippers, cameras, desks, and objects, aiming at a streamlined collection of physical manipulation data and evaluation of the learned skills for hardware benchmarking. We provide a detailed design of the robot cell with readily available parts to build the experiment environment that can host a wide range of robotic hardware commonly adopted for robot learning. We also propose a hierarchical pipeline of software integration, including localization, recognition, grasp planning, and motion planning, to streamline learning-based robot control, data collection, and experiment validation towards shareability and reproducibility. We present benchmarking results of the DeepClaw system for a baseline Tic-Tac-Toe task, a bin-clearing task, and a jigsaw puzzle task using three sets of standard robotic hardware. Our results show that tasks defined in DeepClaw can be easily reproduced on three robot cells. Under the same task setup, the differences in robotic hardware used will present a non-negligible impact on the performance metrics of robot learning. All design layouts and codes are hosted on Github for open access.
Open AccessDeepClaw: A Robotic Hardware Benchmarking Platform for Learning Object Manipulation
We present DeepClaw as a reconfigurable benchmark of robotic hardware and task hierarchy for robot learning. The DeepClaw benchmark aims at a mechatronics perspective of the robot learning problem, which features a minimum design of robot cell that can be easily reconfigured to host robot hardware from various vendors, including manipulators, grippers, cameras, desks, and objects, aiming at a streamlined collection of physical manipulation data and evaluation of the learned skills for hardware benchmarking. We provide a detailed design of the robot cell with readily available parts to build the experiment environment that can host a wide range of robotic hardware commonly adopted for robot learning. We also propose a hierarchical pipeline of software integration, including localization, recognition, grasp planning, and motion planning, to streamline learning-based robot control, data collection, and experiment validation towards shareability and reproducibility. We present benchmarking results of the DeepClaw system for a baseline Tic-Tac-Toe task, a bin-clearing task, and a jigsaw puzzle task using three sets of standard robotic hardware. Our results show that tasks defined in DeepClaw can be easily reproduced on three robot cells. Under the same task setup, the differences in robotic hardware used will present a non-negligible impact on the performance metrics of robot learning. All design layouts and codes are hosted on Github for open access.