Research Assistant, Construction Automation and Robotics Lab

Project: Automatic Alignment of Mesh model and 3D point cloud model for Building Infrastructures
• Implemented Visual SLAM algorithm (ORB SLAM) for 3D reconstruction using 360 degree camera video
• Developed the camera pose estimation, CALTag and ArUCO marker detection algorithm for BIM registration

Project: Real- Time Autonomous Welding System
• Implemented BiseNet, PIDNet models for segmentation, achieving 156fps inference speed with mean IoU of 96%
• Accelerated model inference by 40% using TensorRT and deployed on edge devices (Jetson Nano)
• Implemented LiDAR based obstacle avoidance and autonomous navigation algorithm

Graduate Researcher, Real-Time Intelligent Systems Lab

Project: Automated Bar Hanging with Robotic Arm
• Optimized the MoveIt package in ROS for the robotic arm’s motion planning to accurately pick and hang bars
• Processed RGBD camera-based point cloud data using PCL and implemented DBSCAN for object matching
• Employed Docker for containerization streamlining deployment and project reproducibility

Research Assistant, Neuromuscular Rehabilitation Lab

Project: FSM based Robotic Prosthetic Leg
• Implemented PID algorithm for knee and ankle angle control using Beckhoff PLC
• Executed 50+ benchtop tests, including Torque Response testing and Joint Angle Tracking testing
• Designed and conducted bandwidth tests to investigate system response across a range of frequencies

Graduate Researcher, Construction Automation and Robotics Lab

Project: Real- Time Autonomous Welding System
• Wrote production level robotic arm code for pose estimation and control based on weld plates detection
• Trained the U-Net model for semantic segmentation achieving a mean IOU of 94%
• Labelled and compiled the welding joints dataset, built custom preprocessing pipelines for DNN model input
• Modelled a custom mount using CAD software and 3D printed it to securely hold the welding torch and vision system

Engineer, Larsen & Toubro Technology Services

• Developed traffic-light detection algorithm using YOLO and lane detection using Faster CNN
• Led a team of 10 apprentices for annotation pipeline development ensuring high-quality annotations
• Strategically optimized project workflow resulting in 2 months early completion, significantly increasing profit

North Carolina State University, Raleigh, USA

Masters in Science, Electrical Engineering specialization in Machine Learning and Computer Vision

VSS University of Technology, Burla, India

Bachelor of Technology, Electronics & Telecommunication Engineering