Master Thesis

• SocialNav-Transformer: GPS-Free Social Navigation via Local Visual Context

My master's thesis proposes a vision-based social navigation framework that enables autonomous robots to navigate safely in dynamic, GPS-denied environments. The framework leverages deep visual representations and transformer architectures to predict socially compliant trajectories using only onboard camera observations. 

Keywords: Computer Vision • Transformer • Social Navigation • Robotics • Deep Learning 

Project Details: Github repo

IEEE RA L 

• Calibration and Safety Audit of Transformer-Based Trajectory Prediction

Research project associated with my IEEE Robotics and Automation Letters (RA-L) manuscript. The project evaluates uncertainty calibration, reliability, and safety of transformer-based trajectory prediction models for dense crowd navigation. 

Keywords: Trajectory Prediction • AI Safety • Transformer • Autonomous Navigation 

Project Details: Github repo

Journal Publications

• Wang, Zhicheng, Qingzhang Wang, Lingling Shen, Xiaojun Qian, Jianhua Cui, Liwei Zeng, Zewei Ye, and Md Hasanul Kabir. 2024. “Multi-point Navigation Method for Intelligent Inspection Robots.” Journal of Computing and Electronic Information Management 14 (2): 16-19. https://doi.org/10.54097/1zjue56e

Conference Publications

• Mridha, Krishna Dipayan Barua, Meghla Monir Shorna, Hasan Nouman Nouman, Md Hasanul Kabir, and Ajay Vikram Singh. "Credit approval decision using machine learning algorithms." In 2022 10th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions)(ICRITO), pp. 1-6. IEEE, 2022.