Collision Avoidance Robotics Via Meta-Learning (CARML)

Abhiram Iyer; Aravind Mahadevan

Collision Avoidance Robotics Via Meta-Learning (CARML)

Machine Learning 2020-07-20 v1 Artificial Intelligence Robotics Machine Learning

Authors: Abhiram Iyer , Aravind Mahadevan

Abstract

This paper presents an approach to exploring a multi-objective reinforcement learning problem with Model-Agnostic Meta-Learning. The environment we used consists of a 2D vehicle equipped with a LIDAR sensor. The goal of the environment is to reach some pre-determined target location but also effectively avoid any obstacles it may find along its path. We also compare this approach against a baseline TD3 solution that attempts to solve the same problem.

Keywords

multi-view learning autonomous driving reinforcement learning

Cite

@article{arxiv.2007.08616,
  title  = {Collision Avoidance Robotics Via Meta-Learning (CARML)},
  author = {Abhiram Iyer and Aravind Mahadevan},
  journal= {arXiv preprint arXiv:2007.08616},
  year   = {2020}
}

Related papers

View all related →