Mickyas Tamiru Asfaw's Projects
Aerial Gym - IsaacGym Simulator for Aerial Robots
Implement Convolutional Autoencoder in PyTorch with CUDA
This is an example for basic duckebot practice
A more-involved example ROS submission to the AIDO LF1 Challenge, using the Duckietown Lane Following stack.
Sim-real transfer and submission for duckietown
Repository relating to "Contrastive Initial State Buffer for Reinforcement Learning" (ICRA, 2024).
Towards Monocular Vision Based Collision Avoidance Using Deep Reinforcement Learning
TD3-based Deep Reinforcement learning-based collision avoidance of mobile robots in ROS2.
The set of drivers that run on the robot
PyTorch base image for sim-to-real
Develop and implement reinforcement learning for real-world navigation in DuckieTown, optimizing performance and resilience for reliable autonomous movement, backed by interpretable decision-making tools.
Duckietown Learning Experiences
Source code for our paper "Sim-to-real reinforcement learning applied to end-to-end vehicle control"
Deeper Depth Prediction with Fully Convolutional Residual Networks (FCRN)
PyBullet Gymnasium environments for single and multi-agent reinforcement learning of quadcopter control
Config files for my GitHub profile.
My first python library for ducketown
My first DT project for Ducketown
My first ROS-Compatable DTProject for Duckietown