A5071046462- Robotics and Sensor-Based Localization
- Robotic Path Planning Algorithms
- Distributed Control Multi-Agent Systems
- Video Surveillance and Tracking Methods
- Attention Deficit Hyperactivity Disorder
- Photoacoustic and Ultrasonic Imaging
- Advanced Neural Network Applications
- UAV Applications and Optimization
- Distributed Sensor Networks and Detection Algorithms
- Image and Signal Denoising Methods
- 3D Surveying and Cultural Heritage
- Children's Physical and Motor Development
- Advanced Image Fusion Techniques
- Remote Sensing and LiDAR Applications
- 3D Shape Modeling and Analysis
- Imbalanced Data Classification Techniques
- Advanced Image and Video Retrieval Techniques
- Machine Learning and Data Classification
- Target Tracking and Data Fusion in Sensor Networks
- Perfectionism, Procrastination, Anxiety Studies
- Anomaly Detection Techniques and Applications
Ningbo University
2024-2025
China Academy of Engineering Physics
2021-2024
Three dimensional (3D) object detection with an optical camera and light ranging (LiDAR) is essential task in the field of mobile robot autonomous driving. The current 3D method based on deep learning data-hungry. Recently, semi-supervised (SSOD-3D) has emerged as a technique to alleviate shortage labeled samples. However, it still challenging problem for SSOD-3D learn from noisy pseudo labels. In this paper, dynamically filter unreliable labels, we first introduce self-paced SPSL-3D. It...
Sensor configuration, including the sensor selections and their installation locations, serves a crucial role in autonomous driving. A well-designed configuration significantly improves performance upper bound of perception system. However, as leveraging multiple sensors is becoming mainstream setting, existing methods mainly focusing on single-sensor problems are hardly utilized practice. To tackle these issues, we propose novel method based conditional entropy Bayesian theory to evaluate...
Safe and efficient navigation of a robot in high-density dynamic crowd is challenging task. Most existing algorithms need to acquire the full dynamics neighboring humans at all times, making them heavily dependent on complex upper level information state estimation process. Moreover, scene perception modules do not comprehensively model human–robot interaction spatiotemporal dimension, leading frequent freezing collision problems reinforcement learning-based algorithms. To address above...
This study aimed to discuss the performance of machine learning algorithms on outlier detection problem and investigate effect Bayesian Optimization Algorithm (BOA) improving mod-els.Firstly, Classification And Regression Tree (CART), Extreme Gradient Boosting (XGB), Light Machine (LGBM) models were constructed. The trained using default parameters, it was found that XGB LSTM performed much better than CART. secondly, best hyperparameters searched grid search, random search optimization,...
Abstract Navigating in an unknown area safely is counted as the underlying work which can support swarm agents for more complex tasks. When available information of search regions are lacking, make real-time action decisions according to surrounding environments they have perceived. For agent system, connectivity maintenance and collision avoidance both essential. Based on optimal Reciprocal Collision Avoidance (ORCA) algorithm, we proposed a method that provide assistances by spreading...
Image denoising plays a fundamental role in many computer vision and signal processing tasks. Due to the laser active imaging principle of LiDAR, there is significant speckle noise LiDAR intensity image. To remove enhance image quality, new denosing method based on adaptive non-local means (NLM) exponential soft thresholding proposed. First, converted from multiplicative additive by homomorphic transform tranformed noisy decomposed into high-frequency low-frequency coefficients discrete...