Xiaofeng Liu

ORCID: 0000-0003-1310-6739
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Fault Detection and Control Systems
  • Emotion and Mood Recognition
  • EEG and Brain-Computer Interfaces
  • Robot Manipulation and Learning
  • Video Surveillance and Tracking Methods
  • Anomaly Detection Techniques and Applications
  • Social Robot Interaction and HRI
  • Digital Filter Design and Implementation
  • Face recognition and analysis
  • Image Processing Techniques and Applications
  • Machine Fault Diagnosis Techniques
  • Image and Signal Denoising Methods
  • Speech and Audio Processing
  • Advanced Adaptive Filtering Techniques
  • Speech Recognition and Synthesis
  • Color Science and Applications
  • Autonomous Vehicle Technology and Safety
  • Robotic Path Planning Algorithms
  • Gait Recognition and Analysis
  • Visual perception and processing mechanisms
  • Advanced Sensor and Control Systems
  • Traffic Prediction and Management Techniques
  • Indoor and Outdoor Localization Technologies
  • Robotics and Automated Systems
  • Gaze Tracking and Assistive Technology

Hohai University
2016-2025

Shanghai Jiao Tong University
2025

Institute of Pomology
2024

Chinese Academy of Agricultural Sciences
2024

Lanzhou University of Technology
2024

Wuhu Hit Robot Technology Research Institute
2020-2023

Changzhou University
2020-2023

Second Affiliated Hospital of Soochow University
2022-2023

Soochow University
2022-2023

Tianjin University of Technology and Education
2022

This paper addresses trajectory tracking of an omni-directional mobile robot (OMR) with three mecanum wheels and a fully symmetrical configuration. The wheeled outperforms the non-holonomic due to its ability rotate translate independently simultaneously. A kinematics model OMR is established predictive control (MPC) algorithm system constraints designed achieve point stabilization tracking. Simulation results validate accuracy effectiveness proposed MPC controller.

10.3390/app8020231 article EN cc-by Applied Sciences 2018-02-02

Entropy has been a common index to quantify the complexity of time series in variety fields. Here, we introduce increment entropy measure which each is mapped into word two letters, one letter corresponding direction and other magnitude. The Shannon words termed as (IncrEn). Simulations on synthetic data tests epileptic EEG signals have demonstrated its ability detecting abrupt change, regardless energetic (e.g. spikes or bursts) structural changes. computation IncrEn does not make any...

10.3390/e18010022 article EN cc-by Entropy 2016-01-08

To investigate the effects of muscle fatigue on bioinspired robot learning quality in teaching by demonstration (TbD) tasks, this work, we propose to first identify emerging phenomenon human demonstrator analyzing his/her surface Electromyography (sEMG) recordings and then guide curve with knowledge mind. The time‐varying amplitude frequency sequences determining subband sEMG signals have been estimated their dominant values over short time intervals explored as fatigue‐indicating features....

10.1155/2018/4920750 article EN cc-by Complexity 2018-01-01

Wearable inertial motion capture, a new type of capture technology, mainly estimates the human posture in 3-D space through multisensor data fusion. The available method for sensor fusion is usually aided by magnetometers to remove drift error yaw angle estimation, which turn limits their application presence complex magnetic field environment. In this article, an extended Kalman filter (EKF) proposed fuse 9-axis data. Meanwhile, heuristic reduction (HDR) used calibrate accumulated heading...

10.1109/jiot.2021.3119328 article EN IEEE Internet of Things Journal 2021-10-13

Facial expressions are generally recognized based on handcrafted and deep-learning-based features extracted from RGB facial images. However, such recognition methods suffer illumination/pose variations. In particular, they fail to recognize these with weak emotion intensities. this work, we propose a cross-modality attention-based convolutional neural network (CM-CNN) for expression recognition. We extract expression-related complementary images (gray-scale, local binary pattern, depth...

10.1109/tcds.2022.3150019 article EN IEEE Transactions on Cognitive and Developmental Systems 2022-02-09

BACKGROUND: The gut microbiota plays an important role in human health. It is essential to understand how the composition of neonates established. OBJECTIVES: To investigate nature microbial community first feces newborn infants compared with mothers’ placentae and vaginas. METHODS: One infant who was delivered via Cesarean section vaginally. Bar‐coded pyro‐sequencing 16S ribosomal RNA genes used bacterial structure each site. RESULTS: Neonatal both had similar communities, they were...

10.1155/2015/737294 article EN cc-by Canadian Journal of Infectious Diseases and Medical Microbiology 2015-01-01

We propose an anomaly detection approach by learning a generative model of moving pedestrians to guarantee public safety. To resolve the existing challenges in complicated definitions, complex backgrounds, and local occurrence, weighted convolutional autoencoder-long short-term memory network is proposed reconstruct raw data their corresponding optical flow then perform based on reconstruction errors. Unlike equally treating flow, novel two-stream framework take reconstructed as...

10.1109/tcds.2018.2866838 article EN IEEE Transactions on Cognitive and Developmental Systems 2018-08-23

Abstract Forecasting pedestrians' future trajectory in unknown complex environments is essential to autonomous navigation real‐world applications, for example, self‐driving cars and collision warnings. However, modern observed trajectory‐based prediction methods may easily over‐fit or rare scenes because they do not entirely understand the correlations between trajectories. To address over‐fitting problem, an Inverse Reinforcement Learning Scene‐oriented Trajectory Prediction (IRLSOT)...

10.1049/itr2.12172 article EN cc-by IET Intelligent Transport Systems 2022-02-13

For dimensional emotion recognition, electroencephalography (EEG) signals and electrooculogram (EOG) are often combined to improve the performance of classifiers, as each them provides complementary features other. In this article, we combine EEG signal on relevant channels with EOG boost recognition accuracy. We first explore mutual information (MI) all only select emotion-related channels, i.e., more MI retained, since can be degraded by interference between uncorrelated while...

10.1109/thms.2023.3275626 article EN IEEE Transactions on Human-Machine Systems 2023-05-26

Forecasting pedestrian trajectories in dynamic scenes remains a critical problem various applications, such as autonomous driving and socially aware robots. Such forecasting is challenging due to human-human human-object interactions future uncertainties caused by human randomness. Generative model-based methods handle sampling latent variable. However, few studies explored the generation of In this work, we propose trajectory predictor with pseudo Oracle (TPPO), which generative predictor....

10.1109/tsmc.2024.3351859 article EN IEEE Transactions on Systems Man and Cybernetics Systems 2024-01-30

Dps (DNA protection during starvation) proteins, mini-ferritins in the ferritin superfamily, catalyze Fe(2+)/H(2)O(2)/O(2) reactions and make minerals inside protein nanocages to minimize radical oxygen-chemistry (metal/osmotic/temperature/nutrient/oxidant) sometimes confer virulence. Paired proteins Bacillus, rare other bacteria, have 60% sequence identity. To explore functional differences paired Bacilli protein, we measured ferroxidase activity DNA (hydroxyl radical) for dodecamers from...

10.1074/jbc.m601398200 article EN cc-by Journal of Biological Chemistry 2006-07-22

Lots of evidence has indicated that many kinds animals can achieve goal-oriented navigation by spatial cognition and dead reckoning. The geomagnetic field (GF) is a ubiquitous cue for these animals. Inspired the animals, novel long-distance underwater (LDUGN) method presented in this article, which only utilizes declination component ( D) inclination I) GF without any prior knowledge geographical location or map. D I measured high-precision sensors are compared periodically with destination...

10.1109/tcyb.2019.2933397 article EN IEEE Transactions on Cybernetics 2019-08-22

Robust decoding performance is essential for the practical deployment of brain-computer interface (BCI) systems. Existing EEG models often rely on large amounts annotated data collected through specific experimental setups, which fail to address heterogeneity distributions across different domains. This limitation hinders BCI systems from effectively managing complexity and variability real-world data. To overcome these challenges, we propose Synchronized Self-Training Domain Adaptation...

10.1109/jbhi.2025.3525577 article EN IEEE Journal of Biomedical and Health Informatics 2025-01-01

Collisions between the end-effector (EE) and target are inevitable in robotic arm-based non-cooperative spacecraft capture tasks. These collisions can pose negative effects, such as harming arm causing EE to move away from target, significantly increasing risk of mission failure. Existing research demonstrated that implementing active collision control effectively mitigate these effects. This paper proposes a new one-dimensional method named force-feedback compliance resistance (FFCRC)...

10.2514/1.g008451 article EN Journal of Guidance Control and Dynamics 2025-05-01
Coming Soon ...