Author Image

Hi, I am William

William C Francis

Robotics Graduate Student at GRASP Lab, University of Pennsylvania

Greetings! I am a robotics student passionate about the Artificial Intelligence (AI) industry. My main area of focus lies in prediction and machine learning. I also have a lot of experience in robotics perception, such as computer vision and LiDAR. Hope to see you in the future!

Skills

Experiences

1
Graduate Research Assistant
xLab, University of Pennsylvania

Dec 2021 - May 2022, Philadelphia, PA

xLab is a research lab at University of Pennsylvania lead by Prof. Rahul Mangharam targeting at developing critical systems and autonomous systems

Responsibilities:
  • Developed lane segmentation and obstacle avoidance algorithms using Python and PyTorch to improve the autonomous Go-kart’s ability to navigate through complex racing tracks, resulting in a 40% improvement in obstacle avoidance rate
  • Optimized code for speed using C++, resulting in 15% reduction in response time and real-time processing on Jetson AGX
  • Led the Autonomous Go-kart team to a top-5 finish in the Autonomous evGrandPrix and presented at International Conference on Robotics and Automation (ICRA), securing sponsorship from the Autoware Foundation

Machine Learning Intern
Vision Cultura

Dec 2019 - May 2020, Bangalore, India

Vision Cultura is a start-up that works on IoT-based farming solutions which interconnects sensors, cloud and a mobile app

Responsibilities:
  • Redesigned AI-based farming solution utilizing PyTorch and IoT integration, resulting in 18% crop yield increase and 90% labor cost reduction through real-time monitoring of sunshine, water, and nutrient levels for each plant
  • Deployed models using AWS SageMaker and monitored data in real-time using CloudWatch for optimal analysis
  • Upgraded the perception stack using advanced object detection models (YOLO, RetinaNet, Faster R-CNN) leading to 33% boost in sales team productivity, translating to 100K increase in revenue, earning recognition as outstanding intern
  • Worked in a cross-functional with Mechanical and Computervision engineers, Manager and manufacturers to ensure that product specification, performance, durability, safety, quality, cost, and manufacturing requirements are met
2

3
Undergraduate Research Assistant
Center for Industrial Automation (CIA) Lab, Vellore Instiitute of Technology

Jan 2019 - Dec 2019, Chennai, India

CIA Lab is a research lab at Vellore Institute of Technology lead by Prof. Umayal that builds embedded systems for Industrial Automation and Robotics applications

Responsibilities:
  • Engineered TensorFlow-based deep neural network for blink detection in EEG signals, resulting in 25% increase in signal-to-noise ratio for Brain-Controlled Wheelchair with camera-based collision avoidance
  • Implemented adaptive algorithm for personalized blink detection and control sensitivity, leading to enhanced user experience
  • Conducted user testing and gathered feedback to continually improve the prototype, utilizing a user-centered design approach

Embedded Systems Intern
Siemens Gamesa Renewable Energy

May 2018 - Aug 2018, Bangalore, India

Siemens Gamesa is a leader in the renewable energy industry, working to provide the world’s best offshore and onshore wind turbines and services

Responsibilities:
  • Led design and implementation of miniature windmill model using ATtiny85 microcontroller showcasing novel sensor applications
  • Mounted sensors to optimize windmill performance, designed a PCB using Eagle to control turbine and implemented blade angle detection for Heli-hoist
  • Collected and analyzed sensor data using Parallax Data Acquisition
  • Designed and 3D printed a highly detailed and proportionate model of a wind turbine including the turbine shaft, blades, and tower
4

Education

M.S.E. in Robotics, M.S.E in Computer Science
GPA: 3.8 out of 4
Taken Courses
  • ESE 546 Principles of Deep Learning
  • CIS 581 Computer Vision & Computational Photography
  • CIS 520 Machine Learning
  • ESE 650 Learning in Robotics
  • CIS 522 Deep Learning for Data Science
  • ROBO 599 Independent Study - Build & Race a self-driving Go-Kart
  • CIS 680 Advanced Topics in Machine Perception
  • ESE 514 Graph Neural Networks
  • MEAM 520 Introduction to Robotics
  • CIS 700 Neuro-Symbolic AI
  • CIS 523 Ethics in Algorithms
B.Tech in Electrical and Electronics Engineering with a Minor in Computer Science Engineering
GPA: 8.99 out of 10
Taken Courses
  • CSE 4020 Machine Learning
  • EEE 1007 Neural Networks and Fuzzy Logic
  • CSE 2003 Data Structures and Algorithms
  • CSE 1001 Problem Solving and Programmingg
  • CSE 1002 Problem Solving and Object Oriented Programming
  • MAT 2001 Statistic for Engineers
  • MAT 3003 Complex Variables and Partial Differential Equations
  • EEE 1020 Engineering Optimization
  • EEE 4016 Electric Vehicles
  • CSE 1004 Network and Communication
  • CSE 2002 Theory of Computation and Compiler Design
  • CSE 2001 Computer Architecture and Organization
Extracurricular Activities
  • Captain of the university Tennis team, leading the team to win several tournaments
  • Member of the VIT Quiz Club
  • Awarded the Best Delegate in the inter collegiate Model of United Nations debate competition

Projects

Efficient Panoptic Segmentation with SOLOv2
Developer Aug 2022 - Dec 2022

Improved state-of-the-art EfficientPS model by replacing its instance head (Mask RCNN) with SOLOv2.

Machine Learning for EEG Prosthetic Arm Control
Owner Aug 2021 - Dec 2021

Benchmarking 9 ML algorithms for identifying hand motions from EEG recordings.

Image Super Resolution with improved ESRGAN
Developer Aug 2021 - Dec 2021

Pytorch Implementation of proposed improvements to the state-of-the-art image super resolution model ESRGAN.

Blind Motion Deblurring using deep learning
Developer Aug 2021 - Dec 2021

A novel approach to blind motion deblurring - converting a non-blind method (Weiner Deconvolution) to a blind method using Deep Learning

Robot Localization using PF-LSTM in HabitatAI
Developer Jan 2022 - May 2022

Robot localization using LSTM in HabitatAI simulator, employing particle filter as the embedded algorithmic prior.

Handwritten Digit Recognition using VAE
Owner Aug 2021 - Dec 2021

Variational Autoencoder that learns from MNIST dataset and generates altered handwritten digits.

Particle Filter SLAM for Humanoid Robot
Owner Jan 2022 - May 2022

Indoor SLAM using an IMU and a LiDAR sensor of a humanoid called THOR.

YOLOv3 from scratch
Owner Aug 2022 - Dec 2022

Bottom-up implementation of YOLOv3 for Object Detection.

SOLO and FPN from scratch
Owner Aug 2022 - Dec 2022

pyTorch implementation of Segmenting Objects by LOcation (SOLO) and FPN for COCO dataset.

RNN language model from scratch
Owner Aug 2021 - Dec 2021

Recurrent Neural Network trained on Leo Tolstoy’s War and Peace to imitate his language style.

Email spam classifier using SVM
Developer Mar 2019 - May 2019

Detects spam emails with 98.7% accuracy using Support Vector Machines

Neural network from scratch
Developer Aug 2021 - Dec 2021

Numpy implementation of a neural network for digit recognition

EKF and UKF for 3D orientation tracking
Developer Jan 2022 - May 2022

Robot orientation tracking using EKF and UKF on VICON motion capture system