LAKSHMAN KUMAR

• Use the Crazy Flie Nano UAV platform to simulate the homing behavior of bees

• Created the hardware and simulation test bed for the drone to perform deep-learning tests using NVIDIA Jetson TX1

• Implemented Q Learning, Deep Q Network and Deep Deterministic Policy Gradient algorithms for position control of Crazyflie and proved that they can perform better than classical controllers, like PID, without modeling the dynamics

• Calibrated the cameras and rectified the video feed

• Derived Odometry of the drone using Optical Flow Technique

• Utilized libraries such as ROS, OpenCV, Keras, Tensorflow & Theano and Programmed using C++ & Python

• Involved in development, integration, and validation of 3D robot perception for state of the art autonomous walking legged robots

• Built modules for performing Visual SLAM, Obstacle Avoidance, Object Tracking, Face/Hand Tracking, Sensor Fusion and Pose Estimation

• Identified and integrated diverse advanced perception sensors

• Used libraries such as ROS, PCL, DLIB, OpenCV, OpenMP and programming languages such as C++ & Python

• Assisted with the development of registration and classification database for Face Recognition

• Oversaw the end to end development process of android applications and websites

• Co-developed android applications namely Pocket Counselor ( 10K+ downloads), Pocket Vault, Pocket Tic Tac Toe

• Proposed innovative ideas based on which applications were developed

• Independently wrote the description and privacy policy of the products

• Provided Customer Support