Author name cluster

Maohan Liang

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

6 papers

1 author row

AAAI Conference 2026 Conference Paper

Realistic Curriculum Reinforcement Learning for Autonomous and Sustainable Marine Vessel Navigation

Xiaocai Zhang
Zhe Xiao
Maohan Liang
Tao Liu
Haijiang Li
Wenbin Zhang

Sustainability is becoming increasingly critical in the maritime transport, encompassing both environmental and social impacts, such as Greenhouse Gas (GHG) emissions and navigational safety. Traditional vessel navigation heavily relies on human experience, often lacking autonomy and emission awareness, and is prone to human errors that may compromise safety. In this paper, we propose a Curriculum Reinforcement Learning (CRL) framework integrated with a realistic, data-driven marine simulation environment and a machine learning-based fuel consumption prediction module. The simulation environment is constructed using real-world vessel movement data and enhanced with a Diffusion Model to simulate dynamic maritime conditions. Vessel fuel consumption is estimated using historical operational data and learning-based regression. The surrounding environment is represented as image-based inputs to capture spatial complexity. We design a lightweight, policy-based CRL agent with a comprehensive reward mechanism that considers safety, emissions, timeliness, and goal completion. This framework effectively handles complex tasks progressively while ensuring stable and efficient learning in continuous action spaces. We validate the proposed approach in a sea area of the Indian Ocean, demonstrating its efficacy in enabling sustainable and safe vessel navigation.

PDF Details DOI

EAAI Journal 2025 Journal Article

Big-data-driven vessel destination prediction for smart port management

Jin Chen
Qiang Zhang
Maohan Liang
Chang Peng
Chen Chen

The accurate prediction of vessel destinations is crucial for enhancing maritime traffic efficiency, optimizing port management, and improving regional economic analysis. However, destination information in Automatic Identification System (AIS) data is often missing or inaccurate, which undermines the reliability of maritime analytic. Traditional vessel destination prediction methods primarily focus on measuring trajectory similarities, which results in high computational complexity. This study develops a deep learning approach to vessel destination prediction by transforming the problem into an image classification task. Rasterized images of historical destination ports and vessel trajectories are generated, incorporating AIS data within a fixed spatial context. A multi-scale residual convolutional network is constructed to extract relevant trajectory and port distribution features. To enhance the representation of trajectory endpoints, which are critical for predicting the destination port, a multi-attention mechanism is introduced. This mechanism increases the learning weight assigned to endpoint features, improving prediction accuracy. Finally, a classification network predicts the destination port based on the extracted features. The performance of the proposed method is evaluated using AIS data from the Denmark Strait. Experimental results demonstrate that the model outperforms existing methods, highlighting its potential for applications in smart port management and maritime traffic optimization.

Details DOI

JBHI Journal 2025 Journal Article

Identifying Acute Thoracolumbar Vertebral Compression Fractures From Low-Quality Small-Sample X-Ray Images: A Transfer Learning-Based Approach

Yilin Wang
Weijun Li
Siyu Chen
Yang Yang
Aidi Fan
Chenhao Lei
Yuhui Kou
Na Han

Timely and accurate diagnosis of acute thoracolumbar vertebral compression fractures in X-ray images is critical for initiating prompt and effective treatment, preventing potential neurological damage and long-term disability. Recent advancements in artificial intelligence (AI) have significantly improved medical imaging analysis, providing sophisticated tools to assist clinicians in diagnosing acute thoracolumbar vertebral compression fractures. Nonetheless, detecting these fractures through imaging remains challenging due to the complex overlapping of bony structures in the thoracolumbar region, variability in fracture patterns, and often subtle nature of these injuries. Additionally, the limited availability and sometimes poor quality of medical images further complicate accurate AI-based detection. Addressing these challenges, this study introduces a transfer learning model optimized for recognizing acute thoracolumbar vertebral compression fractures from a small set of low-quality X-ray images. The model starts with a feature extraction model that analyzes multiple texture features of X-ray images. It then employs a Vision Transformer Detector (ViTDet) combined with a faster region-based convolutional neural network (Faster R-CNN) to recognize fractures efficiently. To enhance its performance on small datasets, the model employs a transfer learning approach for training. Extensive experiments with a large dataset of real-world images have shown that this model can effectively recognize acute thoracolumbar vertebral compression fractures from low-quality images, outperforming professionals with specialized knowledge in some cases.

Details DOI

EAAI Journal 2025 Journal Article

Ship anomalous behavior detection based on interval prediction of multiple vessel trajectories

Wentao Liu
Chen Chen
Yaowu Peng
Maohan Liang

In response to the increasing complexity of maritime traffic situations, effective maritime surveillance is essential to ensure the safety of maritime activities. This paper leverages Automatic Identification Systems (AIS) data in conjunction with deep learning techniques and statistical methods to achieve adaptive matching and prediction of multiple vessel trajectories, as well as the identification of anomalous vessel behaviors. Initially, an improved Hausdorff distance is employed to accommodate preprocessed trajectory data, and an enhanced Density-Based Spatial Clustering of Applications with Noise(DBSCAN) algorithm is utilized for cluster analysis, facilitating adaptive multi-vessel trajectory matching. Subsequently, considering the inter-vessel influence zones, a multi-vessel behavior feature prediction is realized based on the completed adaptive vessel trajectory matching. Furthermore, distribution characteristics of trajectory prediction errors are analyzed, and an adapted kernel density estimation method is applied for confidence testing, leading to the proposal of the Unknown Behavioral Expansion- Adaptive Kernel Density Estimate(UBE-AKDE) model for identifying anomalous vessel behaviors. Finally, the algorithms presented are validated using trajectory data from all vessels within and near of the Tianjin Port over a period of five months.

Details DOI

EAAI Journal 2024 Journal Article

Spatio-temporal multi-graph transformer network for joint prediction of multiple vessel trajectories

Ryan Wen Liu
Weixin Zheng
Maohan Liang

The vessel trajectory prediction plays a vital role in guaranteeing traffic safety for unmanned surface vehicles and autonomous surface vessels. By leveraging advanced satellite communication technology, AIS provides massive vessel trajectories, significantly enhancing maritime safety and decision-making. This research proposes a spatio-temporal multi-graph transformer network (ST-MGT), aiming to predict multiple vessel trajectories simultaneously. This innovative model amalgamates the capabilities of graph convolutional networks (GCNs) and transformer models to proficiently address the spatial and temporal interactions amongst vessels. The ST-MGT is comprised of three crucial layers. The temporal transformer layer employs sophisticated temporal transformer and memory mechanisms to discern the intricate temporal correlations between vessel movements. The spatial multi-graph transformer layer constructs a comprehensive multi-graph representation to illuminate spatial correlations between vessels. It incorporates a spatial graph convolutional network and transformer to meticulously understand and interpret the diverse and complex spatial interactions amongst varying vessels. Lastly, the ξ -Regularized LSTM (RegLSTM) layer is implemented for predicting vessel trajectories accurately, based on the unraveled spatio-temporal patterns. Extensive and meticulous experiments reveal that our proposed ST-MGT method transcends other state-of-the-art prediction models in robustness and accuracy. The model’s capability to facilitate multi-vessel and multi-step prediction showcases its immense potential and adaptability in intricate and multifaceted navigation environments, underscoring its practical applicability and significance in enhancing maritime navigational safety.

Details DOI

EAAI Journal 2023 Journal Article

Deep learning-powered vessel traffic flow prediction with spatial-temporal attributes and similarity grouping

Yan Li
Maohan Liang
Huanhuan Li
Zaili Yang
Liang Du
Zhongshuo Chen

Perceiving the future trend of Vessel Traffic Flow (VTF) in advance has great application values in the maritime industry. However, using such big data from the Automatic Identification System (AIS) for accurate VTF prediction remains challenging. Deep training networks can learn valuable features from extensive historical data. This paper proposes a new learning-based prediction network, improved Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) with similarity grouping, including three views. To effectively enable the training network to capture the temporal and periodic (i. e. a spatial attribute) change characteristics of VTF, the CNN and LSTM are employed to compose spatial and temporal views, respectively. Hence, the original one-dimensional data is transformed into a matrix (hour of the day ✕ day) to adapt the input of the proposed methodology. In practical applications, VTF of multiple adjacent target regions need to be predicted simultaneously, and the changes of VTF in different areas may influence each other. To explore their hidden relationships, the similarity grouping view aims to find the target area that exhibits the most similarity with the VTF change trend of the current research area. Furthermore, similar information is combined with the features generated from the other two views to obtain the prediction results. In summary, the new advantage lies in mining the spatiotemporal attributes of data and fusing the similarity information of adjacent regions. Comparative experiments with eleven other methods on realistic VTF datasets show that the proposed method demonstrates superior prediction accuracy and stability performance.

Details DOI