Prithwiraj Paul Email & Phone Number

Wilmington, Massachusetts, United States

Warehouse Robot Control Software Development | Technology Development in Robot Driving capabilities - State Estimation, Motion Planning and Control | Robot Systems and Behavior Design

San Diego, California, United States

As a Teaching Assistant for UCSD's ECE276A course, I aid students in mastering complex concepts in Sensing & Estimation in Robotics. My role involves guiding students through theoretical and practical aspects of Lie groups, Bayesian filtering, and SLAM. I leverage my expertise in linear system theory, probability, computer vision and programming to support students in projects and homework, fostering a deep understanding of sensing and estimation techniques essential in modern robotics.

San Diego, California, United States

Currently working on Vision-guided Autonomous Navigation using Reinforcement LearningDeveloped a Neural Network model for learning autonomous navigation in Ackermann drive vehicles, achieving over 95% accuracy with a reinforcement learning agent trained in a PyBullet and Stable-Baselines3 simulated environment. Executing Sim2Real pipeline for autonomous navigation - Deployed trained RL model using custom C++ ROS node in Nvidia TX2, and benchmarked neural network architectures such as CNN, Transformer, Vision Transformer for image feature analysis.

San Diego, California, United States

Deep Learning for Computer Vision Fast Class-Based Neural Style Transfer (NST)• Developed a custom pipeline for real-time style assignment, achieving 25 fps speed on Tesla T4 GPU with compact 1.64MB model• Utilized Fast-SCNN model for semantic segmentation, with 68.0% Mean Intersection over Union on Cityscape Dataset at 123.5 fps• Performed ablation studies and customized the Fast NST model with an encoder-decoder architecture using PyTorch framework• Used pre-trained VGG16 for improved style transfer with combined content and style loss (Gram matrix) during training• Conducted qualitative study to assess style transfer quality with semantic segmentation on CityScape and PascalVOC images

Boston, Massachusetts, United States

Designed Real-time Safe Trajectory Planning and Control for SymBot - Line Follower Warehouse Automation Robot• Implemented Model Predictive Control with CasADi for trajectory generation, achieving planning times on the scale of seconds and reducing to milliseconds with a Control Barrier Function - Control Lyapunov Function QP framework.• Validated trajectory optimization in C++, yielding a 46% improvement in optimization time and a 10% reduction in turn time, integrated into Symbot’s path planner as an online API.

Bengaluru, Karnataka, India

Project 1: Alleviating water impact loads on Crew Module through active control during parachute descent phase• Developed a 6-DOF trajectory simulator in Simulink for a 5-tonne crewed spacecraft simulating the atmospheric re-entry dynamics• Successfully implemented a robust Bang Bang controller with Proportional Derivative control action with 100% duty cycle• Performed sensitivity studies and achieved a minimum propellant mass of 22 kg (cold gas thruster) for 450 N control force• Obtained results show successful execution of Bang-Bang control to achieve the desired orientation at touchdown but call for a high control force and thrust requirement by running in 100 % duty cycle (simulation done in Simulink and MATLAB)Project 2: Design and Development of Astronaut Training Simulators for India’s first manned mission, Gaganyaan• Configured the Static Mock-up and VR simulator defining the end-to-end requirements between the h/w and s/w modules• Realized an on-board computation tool conveying the spacecraft’s attitude in LVLH reference frame for crew situational awareness• Developed an autonomous quick touchdown point prediction tool during on-orbit spaceflight achieving a prediction error of <1%

Kharagpur, West Bengal, India

As a co-founder of the Hyperloop Research Group at IIT Kharagpur, I led a team of 16 students in designing and building a full-scale working pod prototype for the Hyperloop transportation system.Key achievements of the project include:• Prototyped eddy-current based circular Halbach magnetic levitation module and executed parametric optimization in COMSOL.• Implemented a PID controller and synchronized the speed of BLDC motors, achieving 0 levitation height difference at any instant• Created 2 novel propulsion mechanisms : (i) orientation control of halbach rotor (ii) rotor design modification using EDW propulsion• Achieved 5 times greater propulsive force by tilting rotors than EDW propulsion through electromagnetic simulations in COMSOL

Edmonton, Alberta, Canada

Assessment of energy demand-based GHG mitigation options for Alberta's pulp and paper sector

Bengaluru South, Karnataka, India

Worked on hardware design changes to improve efficiency of vehicle RAC machine in the Bosch Automotive Aftermarket division

Master Of Science - Ms, Electrical And Computer Engineering (Intelligent Systems Robotics And Control), 3.875

Coursework: Sensing and Estimation in Robotics Deep Learning for Computer Vision Planning and Learning in Robotics Robot Manipulation.

Dual Degree, Mechanical Engineering, 9.24

Coursework: Robots and Computer Controlled Machines Soft Computing Systems and Controls Mechatronics Operations Research Finite Element.