Special Sessions

Special Session 2: Multimodal Machine Learning for Intelligent Robotics and Autonomous Systems

Autonomous robots increasingly operate in dynamic and unstructured environments, where reliable perception and decision-making depend on heterogeneous sensory inputs. Advances in machine learning enable the joint interpretation of visible, thermal, depth, LiDAR, inertial, and other sensor data, supporting richer scene understanding than any single modality can provide. However, major challenges remain in cross-modal alignment, learning from limited or noisy data, robust 3D perception, real-time inference, and deployment on resource-constrained platforms. These issues are particularly relevant to unmanned aerial and ground robots, intelligent manufacturing, infrastructure inspection, and human-robot collaboration. This Special Session invites research on machine-learning methods for multimodal perception, localization, mapping, and autonomous decision-making. It welcomes deep learning, self-supervised and generative models, reinforcement learning, and efficient embedded intelligence, emphasizing robustness, experimental validation, and real-world deployment.

Topics of interest for submission include, but are not limited to:

Multimodal representation learning and sensor fusion for robotics

Deep learning for visual, infrared, depth, and LiDAR perception

3D scene understanding, semantic segmentation, and object recognition

Learning-based SLAM, localization, mapping, and reconstruction

Embodied AI and vision-language-action or robot foundation models

Reinforcement learning for autonomous navigation, planning, and control

Collaborative perception, multi-robot systems, and swarm intelligence

Edge AI and real-time machine learning on robotic platforms

Intelligent inspection, anomaly detection, and infrastructure monitoring

Special Session Chairs:

Dr. Yue Ma，School of Robotics, Xi'an Jiaotong-Liverpool University, China

Dr. Yue Ma is an Assistant Professor in the School of Robotics at Xi'an Jiaotong-Liverpool University, China. His research focuses on multimodal perception, infrared thermography, computer vision, deep learning, and intelligent robotic inspection. He works on the integration of heterogeneous sensing technologies for three-dimensional scene understanding, nondestructive evaluation, and structural health monitoring. His interests also include embodied AI and autonomous robotic systems.

The manuscript should be submitted via the submission link (via the Electronic Submission System), or to icmlt_conf@163.com before the submission deadline.