Collision Avoidance Robotics Via Meta-Learning (CARML)
This addresses collision avoidance for robotics, but appears incremental as it applies an existing meta-learning method to a specific domain without reported gains.
The paper tackles the problem of multi-objective reinforcement learning for collision avoidance in robotics, using Model-Agnostic Meta-Learning (MAML) with a 2D vehicle and LIDAR sensor, and compares it to a TD3 baseline, but no concrete results or numbers are provided.
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.