Work
Projects and publications — from research to production robotics.
Projects
5
Self-Driving Car Using Deep Learning on Quanser Qcar
Deep Convolutional Neural Network architecture for autonomous navigation, enabling obstacle detection and avoidance. Trained to differentiate static and dynamic obstacles for collision-free movement, tested on a Quanser Qcar with an Nvidia Jetson TX2.
Clearpath Jackal Frontier Exploration SLAM
Frontier-based exploration using LiDAR in ROS-Melodic Gazebo for autonomous mapping of unknown territories — localization, mapping, and exploration working in tandem with simultaneous localization and mapping (SLAM).
Space Invaders Game using Deep Reinforcement Learning
Deep learning model that learns control policies directly from high-dimensional sensory inputs using reinforcement learning, using RAM state as a compact 128-value environment representation.
Semantic Segmentation
Instance segmentation model trained on COCO and evaluated with PixelLib at Columbus State University's parking lot, using a dashboard-mounted camera feed.
Publications
8
IAES International Journal of Robotics and Automation (IJRA)
A 2D Path-planning Performance Comparison of RRT and RRT* for Unmanned Ground Vehicle
Comparative performance study of RRT and RRT* sampling-based path planners for unmanned ground vehicle navigation.
International Conference on Electrical, Communication and Computer Engineering
Optimizing Mobile Robot Navigation: A Genetic Algorithm and Convolutional Neural Network Approach
Combines genetic algorithm optimization with CNN-based perception to improve mobile robot navigation performance.
IEEE International Conference on Automatic Control and Intelligent Systems
VGG16 and MobileNet Performance Evaluation on Edge Device in Self-Driving Car Technology
Benchmarks VGG16 and MobileNet architectures on edge hardware for real-time self-driving perception workloads.
Intl. Conf. on AI for Smart Community — LNEE vol. 758
EEG Based Brain Controlled RC Car with Attention Level
Brain-computer interface system that drives an RC car using EEG-derived attention levels.
Energy Reports
A Survey on Enhancing Grid Flexibility Through Bidirectional Interactive Electric Vehicle Operations
Survey of vehicle-to-grid strategies for improving power grid flexibility through bidirectional EV charging.
Journal of Renewable and Sustainable Energy
Ancillary Services from Wind and Solar Energy in Modern Power Grids: A Comprehensive Review and Simulation Study
Review and simulation study of ancillary grid services provided by wind and solar generation.
International Journal of Energy Research
Enhancing Load Forecasting Accuracy in Smart Grids: A Parallel Multichannel Network Using 1D CNN and Bi-LSTM
Parallel multichannel deep learning architecture combining 1D CNN and Bi-LSTM for smart grid load forecasting.
Engineering Reports
Time Series Analysis of Solar Power Generation Based on Machine Learning for Efficient Monitoring
Machine learning approach to time-series analysis of solar power generation for efficient plant monitoring.