Worked on Navigation and SLAM for quadruped robot empowered by nav2 stack and slam_toolbox

更多

Developing autonomous navigation system for AMRs for warehouse automation: - Built hardware interface for the motor controllers of diff drive robot - Built custom control module for docking AMR, implemented using state machines based on tf2 from apriltags (attached to the docking station). - Earlier implemented velocity PID controller for custom Docking Module on top of Nav2 - Wrote Behavior Tree plugins for custom behaviors for edge cases of navigation of AMR.

更多

- Implemented and learned state estimation (filter-based), planning and control algorithms in autonomous vehicles. - R&D of learning-based end-to-end Autonomous Driving Systems

更多

- Processed point cloud data using the Point Cloud Library (PCL) and specifically, in removing ground planes using vertical feature removal algorithms such as RANSAC. - Developed perception pipeline for agricultural robotics, including data collection, pre-processing, model training and evaluation. Led the implementation of deep learning-based object detection model and achieved significant improvement in performance through data augmentation by creating synthetic samples using CGAN.

更多

- Worked on all aspects of medical data labeling (tooling), deep learning model training. - Built AI model for skin lesion diagnosis, which involved developing a deep learning model to accurately identify different types of skin lesions.

更多