SACPlanner: Real-World Collision Avoidance with a Soft Actor Critic Local Planner and Polar State Representations

Khaled Nakhleh; Minahil Raza; Mack Tang; Matthew Andrews; Rinu Boney; Ilija Hadžić; Jeongran Lee; Atefeh Mohajeri

ICRA 2023poster10 citations

SACPlanner: Real-World Collision Avoidance with a Soft Actor Critic Local Planner and Polar State Representations

Khaled Nakhleh, Minahil Raza, Mack Tang, Matthew Andrews, Rinu Boney, Ilija Hadžić, Jeongran Lee, Atefeh Mohajeri

Abstract

We study the training performance of ROS local planners based on Reinforcement Learning (RL), and the trajectories they produce on real-world robots. We show that recent enhancements to the Soft Actor Critic (SAC) algorithm such as RAD and DrQ achieve almost perfect training after only 10000 episodes. We also observe that on real-world robots the resulting SACPlanner is more reactive to obstacles than traditional ROS local planners such as DWA.

BibTeX

@inproceedings{icra2023_sacplannerrealwo,
  title = {SACPlanner: Real-World Collision Avoidance with a Soft Actor Critic Local Planner and Polar State Representations},
  author = {Khaled Nakhleh and Minahil Raza and Mack Tang and Matthew Andrews and Rinu Boney and Ilija Hadžić and Jeongran Lee and Atefeh Mohajeri and Karina Palyutina},
  booktitle = {ICRA 2023},
  year = {2023}
}