Using Semantic Information to Improve Generalization of Reinforcement Learning Policies for Autonomous Driving - Archive ouverte HAL Access content directly
Conference Papers Year :

Using Semantic Information to Improve Generalization of Reinforcement Learning Policies for Autonomous Driving

Abstract

The problem of generalization of reinforcement learning policies to new environments is seldom addressed but essential in practical applications. We focus on this problem in an autonomous driving context using the CARLA simulator and first show that semantic information is the key to a good generalization for this task. We then explore and compare different ways to exploit semantic information at training time in order to improve generalization in an unseen environment without finetuning, showing that using semantic segmentation as an auxiliary task is the most efficient approach.
Fichier principal
Vignette du fichier
Carton_Using_Semantic_Information_to_Improve_Generalization_of_Reinforcement_Learning_Policies_WACVW_2021_paper.pdf (973.5 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03110285 , version 1 (14-01-2021)

Identifiers

  • HAL Id : hal-03110285 , version 1

Cite

Florence Carton, David Filliat, Jaonary Rabarisoa, Quoc Cuong Pham. Using Semantic Information to Improve Generalization of Reinforcement Learning Policies for Autonomous Driving. IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Workshops, Jan 2021, Hawaii (on line), United States. ⟨hal-03110285⟩
122 View
263 Download

Share

Gmail Facebook Twitter LinkedIn More