Skip to content
View JhoelWit's full-sized avatar

Highlights

  • Pro

Block or report JhoelWit

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
JhoelWit/README.md

Hi there πŸ‘‹ My name is Jhoel Witter

I'm an aspiring roboticist with a passion for robotics, machine learning, artificial intelligence, and sim-to-real practices. I'm currently interested in working on meaningful and engaging deep learning projects as a robotics software engineer, and growing with the field.

Academic Accomplishments πŸ“–

  • Obtained BS in mechanical engineering from The University at Buffalo βœ”οΈ
  • Obtained MS in robotics specializing in machine learning and artificial intelligence from The University at Buffalo βœ”οΈ

Collaborative Projects πŸ‘¨β€πŸ”¬πŸ‘©β€πŸ”¬

  • Designing an urban air mobility air traffic controller agent for vertiport traffic management:
  • Implementing priority surveillance modes for a scalable coverage and path planning algorithm:
  • Developing a multi-agent dimensionality reduction agent which uses a CNN to effectively downsample observation data:
  • Designing a forex trading agent which takes in multiple timeframes worth of data to make live trades:
  • Creating a natural language processing (NLP) model for integration with messaging applications. To do this, we are using the three step process from OpenAI's InstructGPT and using the openwebtext and twitter unsupervised datasets.

Personal Projects πŸ‘¨β€πŸ”¬

  • Assembling a Petoi Bittle robot, and using deep rl to teach it how to walk with the help of Nvidia's Isaam-Gym. The final model will be applied to the robot using sim-to-real practices:
  • Building a chess engine from scratch and training a multi-agent RL chess agent:
  • Modeling, 3D printing and assembling a 6 DOF arm from scratch, and using a kinect RGB-D camera along with ROS MoveIt! to control it for tasks:

Published Papers πŸ“°

Pinned Loading

  1. Chess_Robot Chess_Robot Public

    A reinforcement learning chess AI

    Python 2

  2. UAM_RL UAM_RL Public

    Designing a reinforcement learning air traffic controller agent for use in an Unreal Engine simulator with an Airsim client.

    Python 5

  3. CNN_Downsampler CNN_Downsampler Public

    A reinforcement learning agent which uses a CNN policy network to effectively downsample observations.

    Python

  4. SCoPP SCoPP Public

    Forked from adamslab-ub/SCoPP

    Scalable Coverage Path Planning of Multi-Robot Teams for Monitoring Non-Convex Areas

    Python

  5. CNN_Cats_Dogs CNN_Cats_Dogs Public

    My first CNN. This model will be trained to differentiate a dataset of cats and dogs.

    Python

  6. bittle-ros-suite bittle-ros-suite Public

    A repository to practice ROS implementations on a Petoi Bittle robot with accesories.

    CMake 1