Kalman Filter for Motion Tracking
Published:
Using a 2D Kalman Filter to obtain smooth motion estimates from noisy sensor data.
Posts by Collection
portfolio
Bayesian Regression for Noisy Sensor Data
Published:
Bayesian linear regression to model uncertainty in sensor-based predictions.
Classifying Fashion-MNIST Images with CNNs
Published:
CNN classifier for Fashion-MNIST achieving around 90.66% test accuracy.
Attention-Enhanced Spatio-Temporal GCN for Air Quality Prediction
Published:
AE-STGCN model combining spatial–temporal attention with graph convolutions for air quality forecasting.
Video-based Obstacle Avoidance using Deep Reinforcement Learning
Published:
Research project developing an intelligent obstacle-avoidance system for autonomous mobile robots using real-time video input. YOLO-based perception is combined with DQN/Double-DQN/PPO policies in ROS/Gazebo to achieve robust collision-free navigation in dynamic environments.
publications
Credit Approval Decision Using Machine Learning Algorithms
Published in 2022 10th International Conference on Reliability, Infocom Technologies and Optimization (ICRITO), 2022
This work evaluates several machine learning algorithms for automated credit approval and compares their performance in terms of accuracy, reliability, and robustness. The results show that ML-based decision systems can significantly improve credit screening efficiency and decision consistency for financial institutions.
Recommended citation: Md. Hasanul Kabir et al., "Credit Approval Decision Using Machine Learning Algorithms," Proceedings of the 2022 10th International Conference on Reliability, Infocom Technologies and Optimization (ICRITO), IEEE, 2022.
Download Paper
Multi-point Navigation Method for Intelligent Inspection Robots
Published in Journal of Computing & Electronic Information Management, 2024
This paper proposes an efficient multi-point navigation method for intelligent inspection robots, focusing on path optimisation, task scheduling, and real-time interaction with the environment. The approach improves navigation accuracy and reduces traversal time in complex inspection scenarios, and was validated on an intelligent inspection robot platform.
Recommended citation: Md. Hasanul Kabir et al., "Multi-point Navigation Method for Intelligent Inspection Robots," Journal of Computing & Electronic Information Management, 2024.
Download Paper