# Simulation-based reinforcement learning for real-world autonomous driving

Source: [Simulation-based reinforcement learning for real-world autonomous driving](<https://www.sota2.com/research/paper/68fa72b91e209131cea582b1>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T02:32:50.348Z

Authors: B{\\l}a\\.zej Osi\\'nski, Adam Jakubowski, Piotr Mi{\\l}o\\'s, Pawe{\\l} Zi\\k{e}cina, Christopher Galias, Silviu Homoceanu, Henryk Michalewski

Published: 2019-11-29T00:08:58.000Z

## Abstract

We use reinforcement learning in simulation to obtain a driving system controlling a full-size real-world vehicle. The driving policy takes RGB images from a single camera and their semantic segmentation as input. We use mostly synthetic data, with labelled real-world data appearing only in the training of the segmentation network.   Using reinforcement learning in simulation and synthetic data is motivated by lowering costs and engineering effort.   In real-world experiments we confirm that we achieved successful sim-to-real policy transfer. Based on the extensive evaluation, we analyze how design decisions about perception, control, and training impact the real-world performance.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1911.12905>)

[PDF](<https://arxiv.org/pdf/1911.12905.pdf>)